Nothing Special   »   [go: up one dir, main page]

WO2019051963A1 - 基于全回路重构仿真的工业控制回路性能评价方法及装置 - Google Patents

基于全回路重构仿真的工业控制回路性能评价方法及装置 Download PDF

Info

Publication number
WO2019051963A1
WO2019051963A1 PCT/CN2017/109548 CN2017109548W WO2019051963A1 WO 2019051963 A1 WO2019051963 A1 WO 2019051963A1 CN 2017109548 W CN2017109548 W CN 2017109548W WO 2019051963 A1 WO2019051963 A1 WO 2019051963A1
Authority
WO
WIPO (PCT)
Prior art keywords
control
reconstructed
control loop
performance
loop
Prior art date
Application number
PCT/CN2017/109548
Other languages
English (en)
French (fr)
Inventor
王建东
杨子江
Original Assignee
山东科技大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 山东科技大学 filed Critical 山东科技大学
Priority to US16/076,440 priority Critical patent/US10611025B2/en
Publication of WO2019051963A1 publication Critical patent/WO2019051963A1/zh

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00—Programme-controlled manipulators
    • B25J9/16—Programme controls
    • B25J9/1628—Programme controls characterised by the control loop
    • B25J9/163—Programme controls characterised by the control loop learning, adaptive, model based, rule based expert control
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B17/00—Systems involving the use of models or simulators of said systems
    • G05B17/02—Systems involving the use of models or simulators of said systems electric
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00—Programme-controlled manipulators
    • B25J9/16—Programme controls
    • B25J9/1628—Programme controls characterised by the control loop
    • B25J9/1653—Programme controls characterised by the control loop parameters identification, estimation, stiffness, accuracy, error analysis
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00—Programme-controlled manipulators
    • B25J9/16—Programme controls
    • B25J9/1674—Programme controls characterised by safety, monitoring, diagnostic
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
    • G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
    • G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
    • G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
    • G05B13/044—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance not using a perturbation signal
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00—Computing arrangements using knowledge-based models
    • G06N5/01—Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N7/00—Computing arrangements based on specific mathematical models

Definitions

  • the invention belongs to the technical field of industrial production control, and in particular relates to an industrial control loop performance evaluation method and device based on full loop reconstruction simulation.
  • the core unit includes nearly one hundred control loops to achieve automatic control of the main variables in the production process. Therefore, the control loop is the key to determine the safe and efficient operation of the process industrial production process. It is very necessary to automatically evaluate the performance of the industrial control loop, find the control loop with poor performance in time, point out the improvement space and technical scheme of the control performance, and make the process The industrial production process maintains good control performance during long-term operation.
  • DCS distributed control system
  • control loop performance evaluation The basic principle of control loop performance evaluation is to find the ideal value that can be achieved by some control performance index under the constraints of some actual conditions, and use it as the reference value to compare with the current value of the performance index to evaluate the dynamics of the control loop.
  • Control performance The technological result of the performance evaluation of the control loop is the performance evaluation method based on minimum variance control proposed by Professor T. Harris of Queen's University in Canada in 1989. It is the most widely accepted and widely used control loop performance evaluation method. However, this method has two main drawbacks:
  • PID Proportional-Integral-Derivative
  • Second, using variance as a performance indicator is more suitable for evaluating the performance of control loops to overcome random noise, and is not suitable for evaluating tracking setpoints and overcoming measurable (deterministic) external disturbances.
  • control module such as filters, piecewise linear functions, dead zones, etc.
  • These control modules are an integral part of the industrial control loop. therefore.
  • the performance evaluation of the control loop it is necessary to consider the important influence of these control modules on the control performance.
  • the prior art only considers the PID controller as a control module, ignoring the other control modules that exist objectively in the industrial control loop, and cannot evaluate the important influence of other control modules on the control performance, resulting in inaccurate, incomplete, or even erroneous performance. Evaluation results.
  • the prior art lacks an effective solution and cannot perform accurate and comprehensive performance evaluation of the control loop of the industrial production process.
  • the present invention provides an industrial control loop performance evaluation method and apparatus based on full loop reconstruction simulation, which effectively solves the problem of industrial control loop performance evaluation, considering both the important influence of the PID controller and the filter and the minute.
  • the important influences of the block linear function and the dead zone link are obtained, so that the control performance evaluation results are consistent with the actual situation of the industrial control loop, and the improvement space of the control performance and the optimization adjustment technical scheme are given.
  • a first object of the present invention is to provide an industrial control loop performance evaluation method based on full loop reconstruction simulation.
  • the present invention adopts the following technical solution:
  • An industrial control loop performance evaluation method based on full loop reconstruction simulation comprising:
  • control module parameters are adjusted according to the control performance index, and the parameters are used to simulate the reconstructed control loop, and the ideal value of the reconstructed performance control index is obtained, and the performance of the control loop is evaluated.
  • the control module includes but is not limited to a PID controller, a filter, a piecewise linear function, and/or a dead zone.
  • the specific steps of determining the correctness of the reconstructed control module are:
  • the controlled value and the set value of the control loop are used as input values of the reconstructed control loop composed of the reconstructed control module, and the output of the reconstructed control loop is obtained as a reconstructed value of the control command of the control loop;
  • the first performance indicator is used to determine the correctness of the reconstructed control module according to the control command of the control loop, the average value of the control command, and the reconstructed value of the control command.
  • the next step is to establish a mathematical model of the controlled object
  • the reconstructed control module has errors, check the construction algorithm of each control module, find out the error and correct, reconstruct the simulation control module, and judge the correctness of the reconstructed control module until the determined control module is determined. Good correctness.
  • the method of establishing a mathematical model of the controlled object includes, but is not limited to, using an autoregressive discrete model, a linear model, or a nonlinear model.
  • the specific steps of optimizing the mathematical model of the controlled object are:
  • An optimization function is established according to the reconstructed entire control loop, and the physical meaning of the optimization function is a degree of fitting between the control command of the entire control loop and the reconstructed value of the controlled quantity and the observed value;
  • the optimization function takes the maximum value to obtain an optimized model of the controlled object.
  • the specific steps are:
  • the optimized model input of the controlled object is obtained as the reconstructed value of the entire control loop control command, and the reconstructed entire control loop output a reconstructed value of the controlled quantity as the entire control loop;
  • Determining the second performance indicator for determining the optimization of the controlled object according to the reconstructed control command of the entire control loop and the weight of the controlled quantity, the average value of the control instruction and the controlled quantity, the control instruction and the reconstructed value of the controlled quantity The correctness of the model.
  • the specific steps of the performance evaluation of the control loop are:
  • the adjusted parameters are used for simulation calculation, and the corresponding reconstruction control performance index is obtained, and the ideal value that the reconstruction control performance index can achieve is obtained;
  • the performance evaluation index of the control loop is obtained, thereby evaluating the control performance of the entire control loop.
  • the reliability of the performance evaluation index of the control loop is clearly determined, and the specific steps are as follows:
  • the stochastic model parameters are generated.
  • the adjusted stochastic model parameters are used for simulation calculation, and the performance evaluation index of the control loop is obtained. Interval.
  • a second object of the present invention is to provide a computer readable storage medium.
  • the present invention adopts the following technical solution:
  • a computer readable storage medium having stored therein a plurality of instructions adapted to be loaded by a processor of a mobile terminal device and to perform the following processing:
  • the parameters of the control module are adjusted according to the control performance index, and the reconstructed control loop is simulated by using the parameter, and the ideal value of the reconstructed performance control index is obtained, and the performance of the control loop is evaluated.
  • a third object of the present invention is to provide a terminal device.
  • the present invention adopts the following technical solution:
  • a terminal device comprising a processor and a computer readable storage medium, the processor for implementing instructions; the computer readable storage medium for storing a plurality of instructions adapted to be loaded by the processor and to perform the following processing:
  • the parameters of the control module are adjusted according to the control performance index, and the reconstructed control loop is simulated by using the parameter, and the ideal value of the reconstructed performance control index is obtained, and the performance of the control loop is evaluated.
  • the present invention can not only evaluate the influence of the PID controller on the control performance, but also evaluate the influence of the control module such as the filter, the piecewise linear function, and the dead zone on the control performance;
  • the second performance index compares the fitness of the controlled quantity with its reconstructed signal, and compares the fitting degree of the control instruction with its reconstructed signal, effectively improving the mathematical model. Estimated accuracy;
  • Figure 1 is a specific flow chart of the method of the present invention
  • Embodiment 2 is a schematic structural view of an industrial control loop in Embodiment 1 of the present invention.
  • FIG. 3 is a schematic diagram showing the principle of a control circuit composed of a reconfiguration control module in Embodiment 1 of the present invention
  • Embodiment 4 is a schematic diagram showing the principle of estimating a mathematical model of a controlled object in Embodiment 1 of the present invention.
  • the invention provides an industrial control loop performance evaluation method and device based on full loop reconstruction simulation, which effectively solves the control performance evaluation problem of the industrial control loop, and considers both the important influence of the PID controller and the filter and the piecewise linear function.
  • the important influences of the modules such as the dead zone link, so as to obtain the control performance evaluation results consistent with the actual situation of the industrial control loop, and give the improvement space of the control performance and the optimization adjustment technical scheme.
  • Embodiment 1 The purpose of Embodiment 1 is to provide an industrial control loop performance evaluation method based on full loop reconstruction simulation.
  • the present invention adopts the following technical solution:
  • An industrial control loop performance evaluation method based on full loop reconstruction simulation comprising:
  • Step (1) performing reconfiguration simulation on the control modules other than the controlled object in the control loop, and judging the correctness of the reconstructed control module;
  • Step (2) establishing a mathematical model of the controlled object, interconnecting with the reconstructed control module, completing the reconstruction of the entire control loop, and optimizing the mathematical model of the controlled object to obtain an optimized model of the controlled object;
  • Step (3) adjust the control module parameters according to the control performance index, use the parameter to perform simulation calculation on the reconstructed control loop, obtain the ideal value of the reconstructed performance control index, and perform control loop performance evaluation.
  • Step (1) Reconstruction of the control module:
  • connection relationship between the control module and the control module is determined according to an actual control loop; the control module includes but is not limited to a PID controller, a filter, a piecewise linear function, and/or a dead zone.
  • the controlled object is a hardware device
  • the control module of the control loop except the controlled object is a software algorithm control module in the DCS, including a PID controller, a filter, and Piecewise linear function and dead zone link, wherein the set value is input between the generated control commands, and the dead zone link, the PID feedback controller and the piecewise linear function are sequentially connected, and the set value is input to the piecewise linear function input.
  • parallel feedforward lead lag filter parallel feedforward lead lag filter. The controlled amount of output is fed back to the input of the deadband link.
  • the reconstruction of the control module is performed by writing a software program in a computer environment, and implementing software algorithms of each module to make it work.
  • the corresponding software modules in the industry control loop have the same function, enabling the reconstruction of software modules in the industrial control loop.
  • the reconstructed control module is shown in Figure 3.
  • the historical data of the control command u(t), the controlled amount y(t), and the set value r(t) are controlled. Based on a performance indicator J u to determine the correctness of the reconstructed control module;
  • the average value of the control command And the reconstructed value of the control instruction Determining the first performance indicator J u is used to determine the correctness of the reconstructed control module:
  • u(t) is the control instruction
  • the value range of the performance index J u is [0, 1]. If the value of J u is close to 1, the effect of the reconstructed control module is good, otherwise the reconstructed control module has an error.
  • the next step is to establish a mathematical model of the controlled object
  • the reconstructed control module has errors, check the construction algorithm of each control module, find out the error and correct, reconstruct the simulation control module, and judge the correctness of the reconstructed control module until the determined control module is determined.
  • the correctness is good, that is, the value of J u is close to 1.
  • Step (2) Estimation of the mathematical model of the controlled object:
  • the controlled object is a physically existing hardware device
  • the input signal is a control command
  • the output signal is a controlled amount.
  • the controlled command in the load control loop of the thermal power unit is the control command applied to the main steam regulating valve.
  • the controlled quantity is the actual power of the generator set.
  • the controlled objects include the main steam regulating valve, steam turbine, generator and other hardware. Device.
  • the method of establishing a mathematical model of the controlled object includes, but is not limited to, using an autoregressive discrete model, a linear model, or a nonlinear model.
  • the mathematical model of the autoregressive discrete model is used to describe the dynamic change process of the output signal of the controlled object in response to the input signal.
  • y(t) is the mathematical model output of the controlled object
  • u(t) is the mathematical model input of the controlled object
  • A(q) and B(q) are the mathematical model polynomials of the controlled object:
  • the estimated parameters are needed.
  • An optimization function is established according to the reconstructed entire control loop, and the physical meaning of the optimization function is a degree of fitting between the control command of the entire control loop and the reconstructed value of the controlled quantity and the observed value;
  • the optimization function takes the maximum value to obtain an optimized model of the controlled object.
  • Figure 4 shows the entire control loop for reconstruction.
  • a new parameter estimation method is proposed.
  • the main idea is to reconstruct the entire control loop based on step (1) and establish a new optimization function.
  • the physical meaning is the weight of the control command and the controlled quantity.
  • the degree of fit between the constructed value and the observed value is obtained by making the optimization function take the maximum value to obtain a mathematical model of the controlled object.
  • the mathematical expression of the parameter estimation method is:
  • u(t) is the control instruction
  • y(t) is the controlled amount.
  • the historical data observations of the set value r(t) are used as the input of the reconstructed entire control loop, and the reconstructed software module and the mathematical model of the controlled object are used to calculate the weight of the control command and the controlled quantity. Construction value.
  • the invention is not limited to a specific technical method to solve the optimization problem, and various optimization methods such as grid search algorithm, least squares algorithm and genetic optimization algorithm can be used to solve the optimization problem. .
  • the optimized model input of the controlled object is obtained as the reconstructed value of the entire control loop control instruction.
  • the reconstructed entire control loop output as the reconstructed value of the controlled amount of the entire control loop
  • the second performance index J y, u is determined to determine the correctness of the optimized model of the controlled object.
  • the second performance index J y, u has a value range of [0, 1], such as J y, u is close to 1, ie with
  • the two reconstructed signals are consistent with the historical data observations of u(t) and y(t), indicating that the mathematical model of the controlled object is of good quality. Otherwise, the mathematical model of the controlled object is of poor quality and needs to find new ones.
  • Model parameters that provide model quality. It should be noted that although With step (1) They are called the reconstructed values of the control instructions, but the calculation methods of the two are significantly different.
  • control loop performance evaluation The specific steps of the control loop performance evaluation are:
  • Step (3-1) selecting a certain control performance indicator to adjust the parameters of the control module; in this embodiment, selecting a control performance index such as an absolute value of the control error, adjusting a PID controller, a filter, a piecewise linear function,
  • the parameters of the module such as the dead zone link are recorded as ⁇ C,i .
  • ⁇ C,i is used in the reconstructed control loop to perform simulation calculation, and the control performance index corresponding to ⁇ C,i is obtained, and the ideal value that the control performance index can reach is found.
  • Step (3-3) comparing the current control performance index with the ideal value of the reconstruction control performance index to obtain a performance evaluation index of the control loop, thereby evaluating the performance of the entire control loop:
  • IAE Optimal is the ideal value for controlling performance indicators: Here It is a controlled quantity reconstruction value corresponding to the adjustment parameter ⁇ C,i . The calculation process is: in the reconstructed control loop, the i-th control loop parameter ⁇ C,i is used to simulate the reconstructed control loop. Calculate the controlled amount. Since IAE Optimal ⁇ IAE Actual , the value range of ⁇ is [0, 1]. If ⁇ is close to 1, it indicates that the current control performance is good, and the potential space for improving performance is small. Otherwise, the control parameters can be greatly improved by adjusting the parameters of the control loop module such as PID controller parameters.
  • the performance evaluation index ⁇ is based on the mathematical model of the controlled object, it is calculated by the full loop reconstruction simulation. Therefore, ⁇ has an uncertainty range, and it is necessary to clarify the reliability of ⁇ .
  • the stochastic model parameters are generated.
  • the adjusted stochastic model parameters are used for simulation calculation, and the performance evaluation index of the control loop is obtained. Interval.
  • the model parameter vector is written as
  • the estimated parameter vector is recorded as
  • the statistical distribution is a Gaussian distribution
  • the mean vector is ⁇
  • the covariance matrix is Cov ⁇ .
  • a performance evaluation index ⁇ 0.9
  • invention 2 is to provide a computer readable storage medium.
  • the present invention adopts the following technical solution:
  • a computer readable storage medium having stored therein a plurality of instructions adapted to be loaded by a processor of a mobile terminal device and to perform the following processing:
  • the parameters of the control module are adjusted according to the control performance index, and the reconstructed control loop is simulated by using the parameter, and the ideal value of the reconstructed performance control index is obtained, and the performance of the control loop is evaluated.
  • Embodiment 2 The purpose of Embodiment 2 is to provide a terminal device.
  • the present invention adopts the following technical solution:
  • a terminal device comprising a processor and a computer readable storage medium, the processor for implementing instructions; the computer readable storage medium for storing a plurality of instructions adapted to be loaded by the processor and to perform the following processing:
  • the parameters of the control module are adjusted according to the control performance index, and the reconstructed control loop is simulated by using the parameter, and the ideal value of the reconstructed performance control index is obtained, and the performance of the control loop is evaluated.
  • examples of the computer readable recording medium include magnetic storage media (for example, ROM, RAM, USB, floppy disk, hard disk, etc.), optical recording media (for example, CD-ROM or DVD), and PC interface (for example, PCI, PCI-Expres, WiFi, etc.).
  • magnetic storage media for example, ROM, RAM, USB, floppy disk, hard disk, etc.
  • optical recording media for example, CD-ROM or DVD
  • PC interface for example, PCI, PCI-Expres, WiFi, etc.
  • the present invention can not only evaluate the influence of the PID controller on the control performance, but also evaluate the influence of the control module such as the filter, the piecewise linear function, and the dead zone on the control performance;
  • the second performance index compares the fitness of the controlled quantity with its reconstructed signal, and compares the fitting degree of the control instruction with its reconstructed signal, effectively improving the mathematical model. Estimated accuracy;

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Theoretical Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Software Systems (AREA)
  • Mechanical Engineering (AREA)
  • Robotics (AREA)
  • Mathematical Physics (AREA)
  • Computing Systems (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Health & Medical Sciences (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Optimization (AREA)
  • Mathematical Analysis (AREA)
  • Algebra (AREA)
  • Computational Mathematics (AREA)
  • Computational Linguistics (AREA)
  • Feedback Control In General (AREA)

Abstract

一种基于全回路重构仿真的工业控制回路性能评价方法及装置，该方法包括：对控制回路中除被控对象之外的控制模块一一进行重构仿真，并判断重构的控制模块的正确性；建立被控对象的数学模型，与重构的控制模块相互连接，完成整个控制回路的重构，并对被控对象的数学模型进行优化，得到被控对象的优化模型；根据控制性能指标调整控制模块参数，使用该参数对重构的控制回路进行仿真计算，得到重构的性能控制指标的理想值，进行控制回路性能评价。对工业控制回路进行重构，以此为基础，评价PID控制器、滤波器、分段线性函数、死区环节等控制模块对控制性能的影响，指出控制性能的改进空间和相应的技术方案。

Description

基于全回路重构仿真的工业控制回路性能评价方法及装置 技术领域
本发明属于工业生产控制的技术领域，尤其涉及一种基于全回路重构仿真的工业控制回路性能评价方法及装置。
背景技术
现代化的电力、石化等流程工业生产过程普遍配置了集散控制系统(DCS：Distributed Control System)，核心单元包括近百个的控制回路，实现对生产过程主要变量的自动控制。因此，控制回路是决定流程工业生产过程的安全高效运行的关键环节，非常有必要自动评价工业控制回路的性能，及时发现性能较差的控制回路，指出控制性能的改进空间和技术方案，使得流程工业生产过程在长期运行中保持良好的控制性能。
控制回路性能评价的基本原理是在一些实际情况的约束条件下，找到某种控制性能指标能够达到的理想值，将其作为基准值，与性能指标的当前值进行比较，从而评价控制回路的动态控制性能。控制回路性能评价的奠基性成果是1989年加拿大Queen's大学T.Harris教授提出的基于最小方差控制的性能评价方法，是最为广泛接受与普遍采用的控制回路性能评价方法。然而，该方法存在两个主要缺点：
第一、未考虑工业界普遍采用的比例-积分-微分(PID：Proportional-Integral-Derivative)控制器的结构对动态控制性能的限制，导致控制性能较好的基于PID控制器的控制回路被误判为性能较差；
第二、采用方差作为性能指标，更适合于评价控制回路克服随机性噪声的性能，而不适用于评价跟踪设定值与克服可测(确定性)外部干扰的性能。
因此，自2004年开始，美国Texas大学T.F.Edgar教授等专家研究PID控制器的结构对控制性能的限制，采用控制误差的绝对值积分等性能指标，通过数值逼近等方法，获得了PID控制器能够达到的控制性能指标的理想值的近似值。然而，在工业应用中，上述现有技术存在一项主要不足之处：
工业控制回路虽然是以PID控制器为核心，但也包含滤波器、分段线性函数、死区环节等控制模块，这些控制模块是工业控制回路必不可少的组成部分。因此。在控制回路性能评价中，需要考虑这些控制模块对控制性能的重要影响。但是， 现有技术只考虑PID控制器这一控制模块，忽略了工业控制回路中客观存在的其它控制模块，无法评价其它控制模块对控制性能的重要影响，导致出现不准确、不全面、甚至错误的性能评价结果。
综上所述，现有技术缺乏有效的解决方案，无法对工业生产过程的控制回路进行准确全面的性能评价。
发明内容
为了解决上述问题，本发明提供一种基于全回路重构仿真的工业控制回路性能评价方法及装置，有效解决工业控制回路性能评价问题，既考虑PID控制器的重要影响，也考虑滤波器、分段线性函数、死区环节等模块的重要影响，从而取得与工业控制回路的实际情况保持一致的控制性能评价结果，给出控制性能的改进空间和优化调整技术方案。
本发明的第一目的是提供一种基于全回路重构仿真的工业控制回路性能评价方法。
为了实现上述目的，本发明采用如下一种技术方案：
一种基于全回路重构仿真的工业控制回路性能评价方法，该方法包括：
对控制回路中除被控对象之外的控制模块一一进行重构仿真，并判断重构的控制模块的正确性；
建立被控对象的数学模型，与重构的控制模块相互连接，完成整个控制回路的重构，并对被控对象的数学模型进行优化，得到被控对象的优化模型；
根据控制性能指标调整控制模块参数，使用该参数对重构的控制回路中进行仿真计算，得到重构的性能控制指标的理想值，进行控制回路性能评价。
根据实际控制回路确定所述控制模块及控制模块之间的连接关系；所述控制模块包括但不限于PID控制器、滤波器、分段线性函数和/或死区环节。
作为进一步的优选方案，所述判断重构的控制模块的正确性的具体步骤为：
根据控制回路的控制指令计算控制指令的平均值；
将控制回路的被控量、设定值作为重构的控制模块组成的重构控制回路的输入值，得到重构控制回路的输出作为控制回路的控制指令的重构值；
根据控制回路的控制指令、控制指令的平均值和控制指令的重构值确定第一性能指标用于判断重构的控制模块的正确性。
作为进一步的优选方案，当判断的重构的控制模块的正确性良好时，执行下一步建立被控对象的数学模型；
否则，重构的控制模块存在错误，检查各个控制模块的构建算法，找出错误并改正，重构仿真控制模块，并判断重构的控制模块的正确性，直至判断的重构的控制模块的正确性良好。
在本发明中，所述建立被控对象的数学模型的方法包括但不限于采用自回归离散模型、线性模型或非线性模型。
作为进一步的优选方案，所述对被控对象的数学模型进行优化的具体步骤为：
根据重构的整个控制回路，建立优化函数，所述优化函数的物理意义为整个控制回路的控制指令和被控量的重构值与观测值之间的拟合度；
使优化函数取最大值，获得被控对象的优化模型。
作为进一步的优选方案，在对被控对象的数学模型进行优化后，判断被控对象的优化模型的正确性，具体步骤为：
确定整个控制回路的控制指令和被控量的权重，两者之和为1；
根据整个控制回路的控制指令计算控制指令的平均值；根据整个控制回路的被控量计算被控量的平均值；
将整个控制回路的被控量、设定值作为重构的整个控制回路的输入值，得到被控对象的优化模型输入作为整个控制回路控制指令的重构值，和重构的整个控制回路输出作为整个控制回路的被控量的重构值；
根据重构的整个控制回路的控制指令和被控量的权重、控制指令和被控量的平均值、控制指令和被控量的重构值确定第二性能指标用于判断被控对象的优化模型的正确性。
作为进一步的优选方案，所述控制回路性能评价的具体步骤为：
选择至少一个控制性能指标对控制模块的参数进行调整；
在重构的整个控制回路中使用调整后的参数进行仿真计算，获得其对应的重构控制性能指标，并获取重构控制性能指标能够达到的理想值；
将选择的控制性能指标与重构控制性能指标的理想值进行比较，得到对控制回路的性能评价指数，从而评价整个控制回路的控制性能。
作为进一步的优选方案，明确得到的控制回路的性能评价指数的可靠性，具体步骤为：
根据被控对象的优化模型参数的统计分布，生成与之匹配的随机模型参数，在重构的整个控制回路中，使用调整后的随机模型参数进行仿真计算，得到控制回路的性能评价指数的置信区间。
本发明的第二目的是提供一种计算机可读存储介质。
为了实现上述目的，本发明采用如下一种技术方案：
一种计算机可读存储介质，其中存储有多条指令，所述指令适于由移动终端设备的处理器加载并执行以下处理：
对控制回路中除被控对象之外的控制模块一一进行重构仿真，并判断重构的控制模块的正确性；
建立被控对象的数学模型，与重构的控制模块相互连接，完成整个控制回路的重构，并对被控对象的数学模型进行优化，得到被控对象的优化模型；
根据控制性能指标调整控制模块参数，使用该参数对重构的控制回路进行仿真计算，得到重构的性能控制指标的理想值，进行控制回路性能评价。
本发明的第三目的是提供一种终端设备。
为了实现上述目的，本发明采用如下一种技术方案：
一种终端设备，包括处理器和计算机可读存储介质，处理器用于实现各指令；计算机可读存储介质用于存储多条指令，所述指令适于由处理器加载并执行以下处理：
对控制回路中除被控对象之外的控制模块一一进行重构仿真，并判断重构的控制模块的正确性；
建立被控对象的数学模型，与重构的控制模块相互连接，完成整个控制回路的重构，并对被控对象的数学模型进行优化，得到被控对象的优化模型；
根据控制性能指标调整控制模块参数，使用该参数对重构的控制回路进行仿真计算，得到重构的性能控制指标的理想值，进行控制回路性能评价。
本发明的有益效果：
通过本发明的一种基于全回路重构仿真的工业控制回路性能评价方法及装置，在计算机环境下，重构了工业控制回路的各个控制模块，对重构的控制回路 进行仿真计算，评价各个控制模块对控制回路性能的影响程度，给出控制性能的改进空间和优化调整技术方案。与现有技术相比，本发明的有益效果包括：
1)扩大了性能评价的对象：本发明不但能够评价PID控制器对控制性能的影响，而且能够评价滤波器、分段线性函数、死区环节等控制模块对控制性能的影响；
2)提高了被控对象数学模型的估计准确性：第二性能指标既对比被控量与其重构信号的拟合度，又对比控制指令与其重构信号的拟合度，有效提高了数学模型的估计准确性；
3)量化了性能评价结果的可靠性：根据模型参数的统计分布，生成与之匹配的随机模型参数，进行重构控制回路的仿真计算，得到了性能评价指数的置信区间。
附图说明
构成本申请的一部分的说明书附图用来提供对本申请的进一步理解，本申请的示意性实施例及其说明用于解释本申请，并不构成对本申请的不当限定。
图1是本发明方法的具体流程图；
图2是本发明实施例1中工业控制回路的结构示意图；
图3是本发明实施例1中重构控制模块组成的控制线路的原理示意图；
图4是本发明实施例1中估计被控对象数学模型的原理示意图。
具体实施方式：
应该指出，以下详细说明都是例示性的，旨在对本申请提供作为进一步的优选方案说明。除非另有指明，本文使用的所有技术和科学术语具有与本申请所属技术领域的普通技术人员通常理解的相同含义。
需要注意的是，这里所使用的术语仅是为了描述具体实施方式，而非意图限制根据本申请的示例性实施方式。如在这里所使用的，除非上下文另外明确指出，否则单数形式也意图包括复数形式，此外，还应当理解的是，当在本说明书中使用术语“包含”和/或“包括”时，其指明存在特征、步骤、操作、器件、组件和/或它们的组合。
在不冲突的情况下，本申请中的实施例及实施例中的特征可以相互组合。下面结合附图与实施例对本发明作进一步说明。
本发明提供一种基于全回路重构仿真的工业控制回路性能评价方法及装置，有效解决工业控制回路的控制性能评价问题，既考虑PID控制器的重要影响，也考虑滤波器、分段线性函数、死区环节等模块的重要影响，从而取得与工业控制回路的实际情况保持一致的控制性能评价结果，给出控制性能的改进空间和优化调整技术方案。
实施例1：
实施例1的目的是提供一种基于全回路重构仿真的工业控制回路性能评价方法。
为了实现上述目的，本发明采用如下一种技术方案：
如图1所示，
一种基于全回路重构仿真的工业控制回路性能评价方法，该方法包括：
步骤(1)：对控制回路中除被控对象之外的控制模块一一进行重构仿真，并判断重构的控制模块的正确性；
步骤(2)：建立被控对象的数学模型，与重构的控制模块相互连接，完成整个控制回路的重构，并对被控对象的数学模型进行优化，得到被控对象的优化模型；
步骤(3)：根据控制性能指标调整控制模块参数，使用该参数对重构的控制回路进行仿真计算，得到重构的性能控制指标的理想值，进行控制回路性能评价。
步骤(1)：控制模块的重构：
根据实际控制回路确定控制模块及控制模块之间的连接关系；所述控制模块包括但不限于PID控制器、滤波器、分段线性函数和/或死区环节。
如图2所示，在本实施例的工业控制回路中，被控对象是硬件装置，控制回路中除被控对象的控制模块为DCS中的软件算法控制模块，包括PID控制器、滤波器、分段线性函数和死区环节，其中由设定值输入至生成控制指令之间，依次连接死区环节、PID反馈控制器和分段线性函数，由设定值输入至分段线性函数输入前，并联前馈的超前滞后滤波器。输出的被控量反馈至死区环节的输入端。
在本实施例中，控制模块(PID控制器、滤波器、分段线性函数和死区环节)的重构采用在计算机环境下编写软件程序，实现各个模块的软件算法，使之与工 业控制回路中相应的软件模块具有相同的功能，从而实现对工业控制回路中软件模块的重新构建。重构后的控制模块如图3所示。
在本实施例中，以控制指令u(t)、被控量y(t)、设定值r(t)的历史数据
Figure PCTCN2017109548-appb-000001
为基础，通过一个性能指标Ju判断重构的控制模块的正确性；
所述判断重构的控制模块的正确性的具体步骤为：
根据控制回路的控制指令u(t)计算控制指令的平均值
Figure PCTCN2017109548-appb-000002
将控制回路的被控量y(t)、设定值r(t)作为重构的控制模块组成的重构控制回路的输入值，得到重构控制回路的输出作为控制回路的控制指令的重构值
Figure PCTCN2017109548-appb-000003
根据控制回路的控制指令u(t)、控制指令的平均值
Figure PCTCN2017109548-appb-000004
和控制指令的重构值
Figure PCTCN2017109548-appb-000005
确定第一性能指标Ju用于判断重构的控制模块的正确性：
Figure PCTCN2017109548-appb-000006
其中，u(t)是控制指令，
Figure PCTCN2017109548-appb-000007
是控制指令的平均值，
Figure PCTCN2017109548-appb-000008
是控制指令的重构值，即是以被控量y(t)、设定值r(t)作为重构的控制模块的输入，对应的重构的控制模块的输出。
在本实施例中，性能指标Ju的取值范围是[0,1]，如Ju取值接近1，则说明重构的控制模块的效果良好，否则重构的控制模块存在错误。
当判断的重构的控制模块的正确性良好时，执行下一步建立被控对象的数学模型；
否则，重构的控制模块存在错误，检查各个控制模块的构建算法，找出错误并改正，重构仿真控制模块，并判断重构的控制模块的正确性，直至判断的重构的控制模块的正确性良好，即Ju取值接近1。
步骤(2)：被控对象的数学模型的估计：
在工业控制回路中，被控对象是物理存在的硬件装置，输入信号是控制指令、输出信号是被控量。
例如，火电机组的负荷控制回路中的被控指令是施加于主蒸汽调节阀的控制指令，被控量是发电机组的实发功率，被控对象包括主蒸汽调节阀、汽轮机、发电机等硬件装置。
在本发明中，所述建立被控对象的数学模型的方法包括但不限于采用自回归离散模型、线性模型或非线性模型。在本实施例中，采用自回归离散模型这一数学模型，描述被控对象这一硬件装置的输出信号响应输入信号的动态变化过程。
自回归离散模型的数学表达式是：
A(q)y(t)＝B(q)u(t)+e(t)，
其中，y(t)是被控对象的数学模型输出、u(t)是被控对象的数学模型输入，A(q)和B(q)是被控对象的数学模型多项式：
Figure PCTCN2017109548-appb-000009
Figure PCTCN2017109548-appb-000010
其中，q-1是移位算子，例如q-1u(t)＝u(t-1)。阶次参数na,nb,nd和模型参数a1,a2,...,ana与b1,b2,...,
Figure PCTCN2017109548-appb-000011
是从控制指令u(t)、被控量y(t)、设定值r(t)的历史数据
Figure PCTCN2017109548-appb-000012
中需要估计的参数。
所述对被控对象的数学模型进行优化的具体步骤为：
根据重构的整个控制回路，建立优化函数，所述优化函数的物理意义为整个控制回路的控制指令和被控量的重构值与观测值之间的拟合度；
使优化函数取最大值，获得被控对象的优化模型。
如图4所示为重构的整个控制回路。
在本实施例中提出一种新的参数估计方法，主要思路是以步骤(1)为基础，重构整个控制回路，建立一个新的优化函数，其物理意义是控制指令与被控量的重构值与观测值之间的拟合度，通过使得优化函数取得最大值，获得被控对象的数学模型。参数估计方法的数学表达式是：
Figure PCTCN2017109548-appb-000013
其中，u(t)是控制指令，
Figure PCTCN2017109548-appb-000014
是控制指令u(t)的平均值，y(t)是被控量，
Figure PCTCN2017109548-appb-000015
是被控量的平均值，wu和wy分别是控制指令和被控量的权重，两者之和为1，即wu+wy＝1；
Figure PCTCN2017109548-appb-000016
与
Figure PCTCN2017109548-appb-000017
分别是以设定值r(t)的历史数据观测值作为重构的整个控制回路的输入，经过重构的软件模块和被控对象的数学模型，计算得到的控制指令和被控量的重构值。
本发明在估计模型参数和阶次参数时，不局限于某种特定的技术方法求解涉及的优化问题，可采用网格搜索算法、最小二乘算法、遗传优化算法等多种技术方法求解优化问题。
在本实施例中，在对被控对象的数学模型进行优化后，判断被控对象的优化模型的正确性，具体步骤为：
确定整个控制回路的控制指令和被控量的权重wu，wy，两者之和为1，即wu+wy＝1；
根据整个控制回路的控制指令u(t)计算控制指令的平均值
Figure PCTCN2017109548-appb-000018
根据整个控制回路的被控量y(t)计算被控量的平均值
Figure PCTCN2017109548-appb-000019
将整个控制回路的被控量y(t)、设定值r(t)作为重构的整个控制回路的输入值，得到被控对象的优化模型输入作为整个控制回路控制指令的重构值
Figure PCTCN2017109548-appb-000020
和重构的整个控制回路输出作为整个控制回路的被控量的重构值
Figure PCTCN2017109548-appb-000021
根据重构的整个控制回路的控制指令和被控量的权重wu，wy、控制指令和被控量的平均值
Figure PCTCN2017109548-appb-000022
控制指令和被控量的重构值
Figure PCTCN2017109548-appb-000023
确定第二性能指标Jy,u用于判断被控对象的优化模型的正确性。
第二性能指标Jy,u的取值范围是[0,1]，如Jy,u接近于1，即
Figure PCTCN2017109548-appb-000024
和
Figure PCTCN2017109548-appb-000025
这两个重构信号分别与u(t)和y(t)的历史数据观测值是一致的，说明被控对象的数学模型品质良好，否则说明被控对象的数学模型品质差，需要寻找新的模型参数，提供模型品质。需要注意的是，虽然
Figure PCTCN2017109548-appb-000026
与步骤(1)中的
Figure PCTCN2017109548-appb-000027
都被称为控制指令的重构值，但两者的计算方法是明显不同的。
步骤(3)：控制性能的评价：
在本实施例中以重构的工业控制回路为基础，提出一种基于随机仿真数据的控制性能评价方法。
所述控制回路性能评价的具体步骤为：
步骤(3-1)：选择某个控制性能指标对控制模块的参数进行调整；在本实施例中选择控制误差绝对值积分等控制性能指标，调整PID控制器、滤波器、分段线性函数、死区环节等模块的参数，记为θC,i。
步骤(3-2)：在重构的整个控制回路中使用调整后的参数进行仿真计算，获得其对应的重构控制性能指标，并获取重构控制性能指标能够达到的理想值；在本实施例中，在重构的控制回路中使用θC,i，进行仿真计算，获得与θC,i对应的控制性能指标，找到控制性能指标能够达到的理想值。
步骤(3-3)：将当前的控制性能指标与重构控制性能指标的理想值进行比较，得到控制回路的性能评价指数，从而对整个控制回路的性能进行评价：
Figure PCTCN2017109548-appb-000028
其中，η为控制误差绝对值积分这一控制性能指标的性能评价指数，IAEActual是控制性能指标的当前值：
Figure PCTCN2017109548-appb-000029
IAEOptimal是控制性能指标的理想值：
Figure PCTCN2017109548-appb-000030
这里
Figure PCTCN2017109548-appb-000031
是与调整参数θC,i对应的被控量重构值，其计算过程是，在重构的控制回路中，使用第i组控制回路参数θC,i，对重构的控制回路进行仿真计算得到的被控量。由于IAEOptimal≤IAEActual，η的取值范围是[0,1]。如η接近于1，则说明当前的控制性能良好，改进性能的潜在空间小，否则说明通过调整PID控制器参数等控制回路模块的参数，可大幅度改进控制性能。
由于性能评价指数η是基于被控对象的数学模型，通过全回路重构仿真计算得到的。因此，η存在不确定性范围，有必要明确η的可靠性。
明确得到的控制回路的性能评价指数的可靠性，具体步骤为：
根据被控对象的优化模型参数的统计分布，生成与之匹配的随机模型参数，在重构的整个控制回路中，使用调整后的随机模型参数进行仿真计算，得到控制回路的性能评价指数的置信区间。
在本实施例中，模型参数向量记为
Figure PCTCN2017109548-appb-000032
估计得到的参数向量记为
Figure PCTCN2017109548-appb-000033
根据系统辨识的渐近性理论结果，
Figure PCTCN2017109548-appb-000034
的统计分布是高斯分布，其均值向量是θ、协方差矩阵是Covθ。根据
Figure PCTCN2017109548-appb-000035
的统计分布，采用Monte-Carlo仿真方法，随机生成位于
Figure PCTCN2017109548-appb-000036
的95％置信区间内的模型参数
Figure PCTCN2017109548-appb-000037
j＝1,2,...,M，将作为被控对象的数学模型，进行全回路重构的仿真计算，得到相应的性能评价指数ηj，以及η的95％置信区间，即ηCI＝[min(ηj),max(ηj)]。例如，某次性能评价指数η＝0.9，95％置信区间是ηCI＝[0.85,0.95]，则说明性能评价结果的置信区间小，性能评价结果的可靠性高。
实施例2：
实施例2的目的是提供一种计算机可读存储介质。
为了实现上述目的，本发明采用如下一种技术方案：
一种计算机可读存储介质，其中存储有多条指令，所述指令适于由移动终端设备的处理器加载并执行以下处理：
对控制回路中除被控对象之外的控制模块一一进行重构仿真，并判断重构的控制模块的正确性；
建立被控对象的数学模型，与重构的控制模块相互连接，完成整个控制回路的重构，并对被控对象的数学模型进行优化，得到被控对象的优化模型；
根据控制性能指标调整控制模块参数，使用该参数对重构的控制回路进行仿真计算，得到重构的性能控制指标的理想值，进行控制回路性能评价。
实施例2：
实施例2的目的是提供一种终端设备。
为了实现上述目的，本发明采用如下一种技术方案：
一种终端设备，包括处理器和计算机可读存储介质，处理器用于实现各指令；计算机可读存储介质用于存储多条指令，所述指令适于由处理器加载并执行以下处理：
对控制回路中除被控对象之外的控制模块一一进行重构仿真，并判断重构的控制模块的正确性；
建立被控对象的数学模型，与重构的控制模块相互连接，完成整个控制回路的重构，并对被控对象的数学模型进行优化，得到被控对象的优化模型；
根据控制性能指标调整控制模块参数，使用该参数对重构的控制回路进行仿真计算，得到重构的性能控制指标的理想值，进行控制回路性能评价。
在本实施例中，计算机可读记录介质的例子包括磁存储介质(例如，ROM，RAM，USB，软盘，硬盘等)、光学记录介质(例如，CD-ROM或DVD)、PC接口(例如，PCI、PCI-Expres、WiFi等)等。然而，本公开的各个方面不限于此。
本发明的有益效果：
通过本发明的一种基于全回路重构仿真的工业控制回路性能评价方法及装置，在计算机环境下，重构了工业控制回路的各个控制模块，对重构的控制回路进行仿真计算，评价各个控制模块对控制回路性能的影响程度，给出控制性能的改进空间和优化调整技术方案。与现有技术相比，本发明的有益效果包括：
1)扩大了性能评价的对象：本发明不但能够评价PID控制器对控制性能的影响，而且能够评价滤波器、分段线性函数、死区环节等控制模块对控制性能的影响；
2)提高了被控对象数学模型的估计准确性：第二性能指标既对比被控量与其重构信号的拟合度，又对比控制指令与其重构信号的拟合度，有效提高了数学模型的估计准确性；
3)量化了性能评价结果的可靠性：根据模型参数的统计分布，生成与之匹配的随机模型参数，进行重构控制回路的仿真计算，得到了性能评价指数的置信区间。
以上所述仅为本申请的优选实施例而已，并不用于限制本申请，对于本领域的技术人员来说，本申请可以有各种更改和变化。凡在本申请的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本申请的保护范围之内。

Claims (10)

  1. 一种基于全回路重构仿真的工业控制回路性能评价方法，其特征是：该方法包括：
    对控制回路中除被控对象之外的控制模块一一进行重构仿真，并判断重构的控制模块的正确性；
    建立被控对象的数学模型，与重构的控制模块相互连接，完成整个控制回路的重构，并对被控对象的数学模型进行优化，得到被控对象的优化模型；
    根据控制性能指标调整控制模块参数，使用该参数对重构的控制回路进行仿真计算，得到重构的性能控制指标的理想值，进行控制回路性能评价。
  2. 如权利要求1所述的一种基于全回路重构仿真的工业控制回路性能评价方法，其特征是：所述控制模块及控制模块之间的连接关系根据实际控制回路确定；所述控制模块包括但不限于PID控制器、滤波器、分段线性函数和/或死区环节。
  3. 如权利要求1所述的一种基于全回路重构仿真的工业控制回路性能评价方法，其特征是：所述判断重构的控制模块的正确性的具体步骤为：
    根据控制回路的控制指令计算控制指令的平均值；
    将控制回路的被控量、设定值作为重构的控制模块组成的重构控制回路的输入值，得到重构控制回路的输出作为控制回路的控制指令的重构值；
    根据控制回路的控制指令、控制指令的平均值和控制指令的重构值确定第一性能指标用于判断重构的控制模块的正确性。
  4. 如权利要求1所述的一种基于全回路重构仿真的工业控制回路性能评价方法，其特征是：当判断的重构的控制模块的正确性良好时，执行下一步建立被控对象的数学模型；
    否则，重构的控制模块存在错误，检查各个控制模块的构建算法，找出错误并改正，重构仿真控制模块，并判断重构的控制模块的正确性，直至判断的重构的控制模块的正确性良好。
  5. 如权利要求1所述的一种基于全回路重构仿真的工业控制回路性能评价方法，其特征是：所述对被控对象的数学模型进行优化的具体步骤为：
    根据重构的整个控制回路，建立优化函数，所述优化函数的物理意义为整个控制回路的控制指令和被控量的重构值与观测值之间的拟合度；
    使优化函数取最大值，获得被控对象的优化模型。
  6. 如权利要求5所述的一种基于全回路重构仿真的工业控制回路性能评价方法，其特征是：在对被控对象的数学模型进行优化后，判断被控对象的优化模型的正确性，具体步骤为：
    确定整个控制回路的控制指令和被控量的权重，两者之和为1；
    根据整个控制回路的控制指令计算控制指令的平均值；根据整个控制回路的被控量计算被控量的平均值；
    将整个控制回路的被控量、设定值作为重构的整个控制回路的输入值，得到被控对象的优化模型输入作为整个控制回路控制指令的重构值，和重构的整个控制回路输出作为整个控制回路的被控量的重构值；
    根据重构的整个控制回路的控制指令和被控量的权重、控制指令和被控量的平均值、控制指令和被控量的重构值确定第二性能指标用于判断被控对象的优化模型的正确性。
  7. 如权利要求1所述的一种基于全回路重构仿真的工业控制回路性能评价方法，其特征是：所述控制回路性能评价的具体步骤为：
    选择至少一个控制性能指标对控制模块的参数进行调整；
    在重构的整个控制回路中使用调整后的参数进行仿真计算，获得其对应的重构控制性能指标，并获取重构控制性能指标能够达到的理想值；
    将选择的控制性能指标与重构控制性能指标的理想值进行比较，得到控制回路的性能评价指数，评价整个控制回路的性能。
  8. 如权利要求7所述的一种基于全回路重构仿真的工业控制回路性能评价方法，其特征是：明确得到的控制回路的性能评价指数的可靠性，具体步骤为：
    根据被控对象的优化模型参数的统计分布，生成与之匹配的随机模型参数，在重构的整个控制回路中，使用调整后的随机模型参数进行仿真计算，得到控制回路的性能评价指数的置信区间。
  9. 一种计算机可读存储介质，其中存储有多条指令，其特征是：所述指令适于由移动终端设备的处理器加载并执行以下处理：
    对控制回路中除被控对象之外的控制模块一一进行重构仿真，并判断重构的控制模块的正确性；
    建立被控对象的数学模型，与重构的控制模块相互连接，完成整个控制回路的重构，并对被控对象的数学模型进行优化，得到被控对象的优化模型；
    根据控制性能指标调整控制模块参数，使用该参数对重构的控制回路进行仿真计算，得到重构的性能控制指标的理想值，进行控制回路性能评价。
  10. 一种终端设备，包括处理器和计算机可读存储介质，处理器用于实现各指令；计算机可读存储介质用于存储多条指令，其特征是：所述指令适于由处理器加载并执行以下处理：
    对控制回路中除被控对象之外的控制模块一一进行重构仿真，并判断重构的控制模块的正确性；
    建立被控对象的数学模型，与重构的控制模块相互连接，完成整个控制回路的重构，并对被控对象的数学模型进行优化，得到被控对象的优化模型；
    根据控制性能指标调整控制模块参数，使用该参数对重构的控制回路进行仿真计算，得到重构的性能控制指标的理想值，进行控制回路性能评价。
PCT/CN2017/109548 2017-09-12 2017-11-06 基于全回路重构仿真的工业控制回路性能评价方法及装置 WO2019051963A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US16/076,440 US10611025B2 (en) 2017-09-12 2017-11-06 Method an device for evaluating performance of industrial control loops based on full loop reconstruction simulations

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710817779.XA CN108121215B (zh) 2017-09-12 2017-09-12 基于全回路重构仿真的工业控制回路性能评价方法及装置
CN201710817779.X 2017-09-12

Publications (1)

Publication Number Publication Date
WO2019051963A1 true WO2019051963A1 (zh) 2019-03-21

Family

ID=62228247

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/109548 WO2019051963A1 (zh) 2017-09-12 2017-11-06 基于全回路重构仿真的工业控制回路性能评价方法及装置

Country Status (3)

Country Link
US (1) US10611025B2 (zh)
CN (1) CN108121215B (zh)
WO (1) WO2019051963A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111983997A (zh) * 2020-08-31 2020-11-24 北京清大华亿科技有限公司 一种基于耦合性分析的控制回路性能监测方法及系统
CN113904933A (zh) * 2021-09-30 2022-01-07 中国船舶重工集团公司第七一九研究所 数字化网络控制系统性能与可靠性同步优化方法

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112083704A (zh) * 2019-06-12 2020-12-15 中国石油化工股份有限公司 化工装置控制回路性能定量评估方法、机器可读存储介质
CN110557390B (zh) * 2019-09-05 2021-08-10 中创智客(苏州)智能科技有限公司 一种分散控制系统与第三方系统的安全通讯方法及系统
CN112621746A (zh) * 2020-12-02 2021-04-09 上海交通大学烟台信息技术研究院 带死区的pid控制方法及机械臂视觉伺服抓取系统
CN114217583B (zh) * 2021-12-10 2024-04-12 北京国控天成科技有限公司 流程工业控制回路自控率、平稳率统计方法及系统
CN117313409B (zh) * 2023-10-17 2024-04-02 国网四川省电力公司电力科学研究院 一种电弧数学模型参数优化方法、系统、终端及介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110288846A1 (en) * 2010-05-21 2011-11-24 Honeywell International Inc. Technique and tool for efficient testing of controllers in development (h-act project)
CN102393627A (zh) * 2011-10-21 2012-03-28 靳其兵 集散系统中控制器参数的优化整定系统和方法
CN102998974A (zh) * 2012-11-28 2013-03-27 上海交通大学 多模型广义预测控制系统及其性能评估方法
CN103309237A (zh) * 2013-06-03 2013-09-18 上海交通大学 基于多模型混合最小方差控制的时变扰动控制系统性能评估方法
CN107085420A (zh) * 2017-05-05 2017-08-22 浙江理工大学 基于最小方差的控制回路性能评估方法及系统

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6122557A (en) * 1997-12-23 2000-09-19 Montell North America Inc. Non-linear model predictive control method for controlling a gas-phase reactor including a rapid noise filter and method therefor
CN104298226A (zh) * 2014-09-28 2015-01-21 中国石油大学(华东) 一种基于状态的控制系统性能评价装置与方法
CN104898605A (zh) * 2015-04-09 2015-09-09 华北电力大学（保定） 一种基于协方差指标的火电机组负荷控制系统性能评价方法
CN105278527B (zh) * 2015-12-01 2017-10-10 中国烟草总公司郑州烟草研究院 一种适用于烟草加工过程单回路控制系统的实时性能评价方法
CN105334738B (zh) * 2015-12-01 2018-10-09 中国烟草总公司郑州烟草研究院 一种适用于烟草加工过程pid控制回路的性能评价方法
CN106774243B (zh) * 2016-12-02 2019-07-30 浙江中控软件技术有限公司 Pid性能评估方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110288846A1 (en) * 2010-05-21 2011-11-24 Honeywell International Inc. Technique and tool for efficient testing of controllers in development (h-act project)
CN102393627A (zh) * 2011-10-21 2012-03-28 靳其兵 集散系统中控制器参数的优化整定系统和方法
CN102998974A (zh) * 2012-11-28 2013-03-27 上海交通大学 多模型广义预测控制系统及其性能评估方法
CN103309237A (zh) * 2013-06-03 2013-09-18 上海交通大学 基于多模型混合最小方差控制的时变扰动控制系统性能评估方法
CN107085420A (zh) * 2017-05-05 2017-08-22 浙江理工大学 基于最小方差的控制回路性能评估方法及系统

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
SUN, JINMING ET AL.: "PID (PID Performance Assessment", CONTROL AND INSTRUMENTS IN CHEMICAL INDUSTRY, vol. 31, no. 5, 31 December 2014 (2014-12-31), pages 21 - 23, ISSN: 1000-3932 *
YU , ZHENPENG ET AL.: "Performance Assessment of Static Lead-Lag Feedforward Controllers for Disturbance Rejection in PID Control Loops", ISA TRANSACTIONS, vol. 64, 4 May 2016 (2016-05-04), pages 67 - 76, XP029731710, ISSN: 0019-0578, DOI: doi:10.1016/j.isatra.2016.04.016 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111983997A (zh) * 2020-08-31 2020-11-24 北京清大华亿科技有限公司 一种基于耦合性分析的控制回路性能监测方法及系统
CN111983997B (zh) * 2020-08-31 2021-07-20 北京清大华亿科技有限公司 一种基于耦合性分析的控制回路性能监测方法及系统
CN113904933A (zh) * 2021-09-30 2022-01-07 中国船舶重工集团公司第七一九研究所 数字化网络控制系统性能与可靠性同步优化方法
CN113904933B (zh) * 2021-09-30 2023-08-25 中国船舶重工集团公司第七一九研究所 数字化网络控制系统性能与可靠性同步优化方法

Also Published As

Publication number Publication date
CN108121215B (zh) 2018-11-16
US20190160661A1 (en) 2019-05-30
US10611025B2 (en) 2020-04-07
CN108121215A (zh) 2018-06-05

Similar Documents

Publication Publication Date Title
WO2019051963A1 (zh) 基于全回路重构仿真的工业控制回路性能评价方法及装置
CN110598929B (zh) 一种风电功率非参数概率区间超短期预测方法
WO2020010629A1 (zh) 过程控制预测模型调整方法、装置和过程控制器
WO2020258826A1 (zh) 基于复杂网络模型的工业设备运维与优化方法及系统
CN116562156B (zh) 一种控制决策模型的训练方法、装置、设备及存储介质
CN113868953B (zh) 工业系统中多机组运行优化方法、装置、系统及存储介质
CN104408317A (zh) 一种基于Bootstrap回声状态网络集成的冶金企业煤气流量区间预测方法
CN116165976A (zh) 生产系统的控制方法、装置、系统、设备及存储介质
JP2021174397A (ja) 制御支援装置、制御支援方法、制御支援プログラム、および制御システム
WO2022267209A1 (zh) 并联阀组控制方法、装置、控制器和存储介质
CN103197692B (zh) 一种流量控制方法、装置及系统
US11287126B2 (en) System and method for controlling operation of boiler
CN108537322A (zh) 神经网络层间激活值量化方法及装置
JP7216566B2 (ja) 情報処理装置、情報処理方法、および情報処理プログラム
WO2022246627A1 (zh) 一种用于控制制冷设备的方法和装置
JP6848710B2 (ja) プラント制御調整装置及び方法
CN114383130B (zh) 锅炉过热汽温度控制方法、装置、设备及存储介质
CN111443602A (zh) 一种基于隐马尔可夫的阀门健康度估计与寿命预测方法
CN114139442A (zh) 基于K-means-LSTM的汽轮机阀门流量特性函数的优化方法
CN108445749B (zh) 一种应用于高阶滑模控制器的参数整定方法
CN117557070B (zh) 能耗优化方法、装置及电子设备
CN110516323A (zh) 一种基于时序分析的压力管道损伤预测方法
Zhou et al. Reliable identification based intelligent PID tuning for long-period process control under different working conditions
Zhou et al. Design System of Refrigerator Drive Control Circuit Based on Deep Learning Algorithm
Chen et al. The improved Fuzzy-Smith compensator based on modeling error for network control system

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17925101

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17925101

Country of ref document: EP

Kind code of ref document: A1