Multi-effect evaporator group based on thermal parameter collaborative optimization and control method thereof

Abstract

The application relates to the technical field of evaporation, in particular to a multi-effect evaporator group based on thermal parameter collaborative optimization and a control method thereof. The method collects thermal parameters of each effect of the multi-effect evaporator group, establishes thermal parameter sequences of each effect in each control period, calculates a coupling risk index and a control response sensitivity of the control period, adaptively adjusts a forgetting factor according to the control response sensitivity, uses an adaptive model predictive control algorithm, and realizes the control of the multi-effect evaporator group according to all thermal parameter sequences of all effects of the multi-effect evaporator group in the same control period. The application can improve the stability of the control of the multi-effect evaporator group.

Description

Technical Field

[0001] This application relates to the field of evaporation technology, specifically to a multi-effect evaporator array and its control method based on synergistic optimization of thermodynamic parameters. Background Technology

[0002] Multi-effect evaporator units are indispensable core thermal separation process equipment in fields such as chemical production, seawater desalination, and zero discharge of high-salinity industrial wastewater. A multi-effect evaporator unit consists of multiple independent evaporation units connected in series to form a single complete system. Each independent evaporation unit is one effect. Single-effect equipment typically features two core structures: a heating chamber and a separation chamber. During operation, the secondary steam generated in the previous effect can be directly used as the heating source for the next effect. Through multi-stage heat energy recycling between effects, the consumption of external heating steam is significantly reduced, meeting the core needs of energy conservation, emission reduction, and green sustainable development in industrial production.

[0003] Currently, the operating parameters of each effect in a multi-effect evaporator are not independently controllable. Key thermodynamic parameters such as effect temperature, operating pressure, tank liquid level, and feed flow rate form a strongly correlated and highly nonlinear coupling constraint relationship based on the internal thermodynamic cycle. The heating steam pressure of the previous effect directly determines the secondary steam saturation temperature, indirectly controlling the heat transfer temperature difference and material evaporation intensity of the next effect. Even a small disturbance in any single operating parameter will be transmitted and amplified step by step through the pressure and temperature gradients between effects, affecting the overall stability of the unit. At present, the industry generally adopts the traditional single-loop PID independent control mode, adjusting the valve opening and operating conditions separately for a single operating variable. This control method does not consider the thermodynamic coupling characteristics between effects. The adjustment operation will disrupt the original thermodynamic balance of the multi-effect evaporator group, causing abnormal fluctuations in the thermodynamic parameters of each effect, which in turn causes severe oscillations in the parameters of the multi-effect evaporator group, directly reducing the overall evaporation heat exchange and material handling efficiency. When the operating condition is severely unbalanced, it will also trigger equipment interlock protection shutdown, affecting the continuous and stable operation of the production line and restricting the high-quality and efficient implementation of industrial evaporation processes. Summary of the Invention

[0004] To address the aforementioned technical problems, this application provides a multi-effect evaporator array and its control method based on synergistic optimization of thermodynamic parameters. The specific technical solution adopted is as follows: In a first aspect, one embodiment of this application provides a control method for a multi-effect evaporator group based on coordinated optimization of thermodynamic parameters, the method comprising the following steps: Thermodynamic parameters of each effect in the multi-effect evaporator group are collected at different collection times, and a sequence of thermodynamic parameters of each effect in each control cycle is established. The thermodynamic parameters include steam temperature, pressure, feed flow rate and liquid level. Based on the sequence of all thermodynamic parameters and the corresponding mutual information entropy of all different effects of the multi-effect evaporator group in the same control cycle, the coupling risk index of the control cycle is calculated. The coupling risk index is used to evaluate the impact of the linear and nonlinear relationships of thermodynamic parameters on the stability of the multi-effect evaporator group. Based on the differences between all thermodynamic parameter sequences of all different effects in the same control cycle of the multi-effect evaporator group, and the coupling risk index of the control cycle, the control response sensitivity of the control cycle is calculated. The control response sensitivity is used to evaluate the degree of synchronization of the responses of each effect in the multi-effect evaporator group. Based on the control response sensitivity of the control cycle, the forgetting factor of the adaptive model predictive control algorithm is adaptively adjusted. Using the adaptive model predictive control algorithm, the control of the multi-effect evaporator group is realized based on the sequence of all thermodynamic parameters of all effects in the same control cycle.

[0005] Furthermore, the method for obtaining the coupling risk index of the control cycle is as follows: Based on the sequence of all thermodynamic parameters of all effects of the multi-effect evaporator group in the same control cycle, the contribution rate of the first principal component in the same control cycle is extracted; In the same control cycle, based on the steam temperature sequence, pressure sequence, feed flow rate sequence, and liquid level sequence of the two effects, the mutual information entropy between the steam temperature, pressure, feed flow rate, and liquid level of the two effects is calculated respectively, and the comprehensive mutual information entropy of the same control cycle is calculated. The weighted sum of the contribution rate of the first principal component of the control cycle and the comprehensive mutual information entropy is denoted as the coupling risk index of the control cycle, where the weighted sum of the contribution rate of the first principal component of the control cycle and the comprehensive mutual information entropy is 1.

[0006] Furthermore, the method for obtaining the contribution rate of the first principal component of the control cycle is as follows: Based on the thermodynamic parameter sequences of all effects of the multi-effect evaporator group in the same control cycle, a thermodynamic parameter matrix for the same control cycle is established, and the contribution rate of the first principal component in the same control cycle is extracted using a principal component analysis algorithm.

[0007] Furthermore, the method for obtaining the comprehensive mutual information entropy of the control cycle is as follows: The arithmetic mean of the four mutual information entropies is denoted as the corresponding thermodynamic parameter mutual information entropy of the two effects. The arithmetic mean of the normalized values ​​of the mutual information entropies of the thermodynamic parameters of all different effects of the multi-effect evaporator group is denoted as the comprehensive mutual information entropy of the same control cycle.

[0008] Furthermore, the method for obtaining the control response sensitivity of the control cycle is as follows: In the same control cycle, the DTW distances of the steam temperature sequence, pressure sequence, feed flow rate sequence and liquid level sequence of the two effects are calculated respectively. The arithmetic mean of the four DTW distances is recorded as the corresponding thermodynamic parameter DTW distance of the two effects. The thermodynamic parameter variation coefficient and the average thermodynamic parameter DTW distance of the control cycle are determined. The normalized value of the coefficient of variation of the thermodynamic parameters of the control cycle is denoted as the characteristic index value of the control cycle. The number 1 is used as the numerator. The product of the characteristic index value of the control cycle and the coupling risk index is denoted as the characteristic product. The normalized value of the average thermodynamic parameter DTW distance of the control cycle is calculated and the sum of the number 1 is calculated. The product of the sum and the characteristic product is used as the denominator. The normalized value of the fraction is denoted as the control response sensitivity of the control cycle.

[0009] Furthermore, the coefficient of variation of the thermodynamic parameters of the control cycle is: the coefficient of variation of the DTW distance of the thermodynamic parameters of all different effects of the multi-effect evaporator group.

[0010] Furthermore, the average thermodynamic parameter DTW distance is: the arithmetic mean of all different effects of the multi-effect evaporator group.

[0011] Furthermore, the specific method for adaptively adjusting the forgetting factor of the adaptive model predictive control algorithm based on the control response sensitivity of the control cycle includes: The product of the difference between the number 1 and the control response sensitivity of the control cycle and the preset adjustment coefficient is recorded as the sensitivity product of the control cycle. The difference between the preset initial forgetting factor and the sensitivity product of the control cycle is used as the adaptive value of the forgetting factor of the adaptive model predictive control algorithm in the control cycle.

[0012] Furthermore, the control method of the multi-effect evaporator group is as follows: based on the control sequence output by the adaptive model predictive control algorithm, the opening degree of each valve of the multi-effect evaporator group is controlled at each sampling time within one minute after the control cycle.

[0013] Secondly, another embodiment of this application provides a multi-effect evaporator group based on thermodynamic parameter co-optimization, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the above-described multi-effect evaporator group control method based on thermodynamic parameter co-optimization.

[0014] The embodiments of this application have at least the following beneficial effects: This application first extracts the linear and nonlinear coupling characteristics of the thermodynamic parameters of each evaporation unit in a multi-effect evaporator group, evaluates the impact of the linear and nonlinear relationships of the thermodynamic parameters on the stability of the multi-effect evaporator group, and obtains the coupling risk index of the control cycle. Since parameter coupling between effects can generate a high risk of oscillation, even if the coupling strength between the thermodynamic parameters of each effect is low, differences in the control actions of each effect can lead to a decrease in evaporator efficiency. Furthermore, the interference of thermodynamic parameters on the control loops of each effect and the differences in control response are quantified to evaluate the synchronization degree of the responses of each effect in the multi-effect evaporator group and obtain the control response sensitivity of the control cycle. Based on... The dynamic changes in the response mode of the multi-effect evaporator group are adaptively adjusted to adjust the forgetting factor of the adaptive model predictive control algorithm, thereby enhancing the adaptability of the adaptive model predictive control algorithm to time-varying characteristics. This effectively suppresses parameter oscillations and improves the operational stability and control accuracy of the multi-effect evaporator group. Finally, the adaptive model predictive control algorithm is used to control the multi-effect evaporator group based on the sequence of all thermodynamic parameters of all effects in the same control cycle. This solves the problem that existing control methods ignore the nonlinear coupling relationship formed between the effects of the multi-effect evaporator group through thermodynamic cycles, which leads to thermodynamic parameter oscillations and a decrease in evaporation efficiency. Attached Figure Description

[0015] Figure 1 This is a flowchart illustrating the steps of a multi-effect evaporator group control method based on thermodynamic parameter co-optimization, as provided in one embodiment of this application. Detailed Implementation

[0016] The following description, in conjunction with the accompanying drawings, details the specific scheme of the multi-effect evaporator group and its control method based on the coordinated optimization of thermodynamic parameters provided in this application.

[0017] Please see Figure 1 The diagram illustrates a flowchart of a multi-effect evaporator group control method based on thermodynamic parameter co-optimization according to an embodiment of this application. The method includes the following steps: Step S001: Collect the thermodynamic parameters of each effect of the multi-effect evaporator group at different collection times, and establish the thermodynamic parameter sequence of each effect in each control cycle. The thermodynamic parameters include steam temperature, pressure, feed flow rate and liquid level.

[0018] Temperature sensors and pressure transmitters are installed at the steam inlet of each effect in the multi-effect evaporator group, electromagnetic flow meters are installed at the feed end, and differential pressure level transmitters are installed in the level chamber. The temperature, pressure, feed flow and level of the corresponding effect are collected by the temperature sensor, pressure transmitter, electromagnetic flow meter and differential pressure level transmitter respectively.

[0019] Preferably, in one embodiment of this application, when collecting data on steam temperature, pressure, feed flow rate, and liquid level, the data sampling frequency is 50Hz, and the control cycle is set to 1 minute. In practical applications, as other implementation methods, the implementer can determine the sampling frequency and control cycle value according to the actual situation, and this application does not impose any special restrictions.

[0020] It should be noted that, for ease of calculation, all steam temperature, pressure, feed flow rate, and liquid level involved in the calculation in this embodiment have undergone data preprocessing to eliminate the influence of dimensions. This embodiment uses the Z-Score standard normalization method to remove dimensions from steam temperature, pressure, feed flow rate, and liquid level. In practical applications, implementers can use other methods such as the maximum-minimum value normalization method for dimension removal, which are not limited here.

[0021] Steam temperature, pressure, feed flow rate, and liquid level are all recorded as thermodynamic parameters. A sequence of thermodynamic parameters for each effect of the multi-effect evaporator group is established for each control cycle.

[0022] It is understood that one effect is an independent evaporation unit of a multi-effect evaporator group; the thermodynamic parameter sequence includes steam temperature sequence, pressure sequence, feed flow rate sequence and liquid level sequence.

[0023] At this point, the thermodynamic parameter sequences of each effect in each control cycle are obtained.

[0024] Step S002: Based on the sequence of all thermodynamic parameters and the corresponding mutual information entropy of all different effects of the multi-effect evaporator group in the same control cycle, calculate the coupling risk index of the control cycle. The coupling risk index is used to evaluate the influence of the linear and nonlinear relationships of thermodynamic parameters on the stability of the multi-effect evaporator group.

[0025] During the control process of the multi-effect evaporator group, there are significant nonlinear coupling characteristics between the thermodynamic parameters of each evaporation unit. Small fluctuations in a single thermodynamic parameter will spread and affect each other through the heat transfer effect of the multi-effect evaporator group, which can easily cause the overall operating parameters of the multi-effect evaporator group to fluctuate continuously, ultimately leading to a reduction in evaporation efficiency. In severe cases, it can even cause unplanned shutdowns of production equipment, affecting the continuous and stable operation of the workshop.

[0026] All thermodynamic parameter sequences of all effects in the multi-effect evaporator group within the same control cycle are arranged sequentially from left to right as column vectors to establish a thermodynamic parameter matrix for the corresponding control cycle. Principal component analysis (PCA) is then used to obtain the contribution rate of the first principal component within the same control cycle.

[0027] In the thermodynamic parameter matrix, this embodiment sets the thermodynamic parameter sequences of the same effect from left to right as steam temperature sequence, pressure sequence, feed flow rate sequence, and liquid level sequence. The column vectors corresponding to each effect of the multi-effect evaporator group are arranged from left to right according to the order of each effect in the multi-effect evaporator group. Taking a multi-effect evaporator group containing 3 effects as an example, the thermodynamic parameter matrix is ​​3000. A 12-dimensional matrix; each row of the thermodynamic parameter matrix corresponds to a sampling time, and each column corresponds to a certain type of thermodynamic parameter sequence for a certain effect. The arrangement of the column vectors of the thermodynamic parameter matrix can be set according to the requirements, and this application does not impose any special restrictions.

[0028] Principal component analysis (PCA) projects high-dimensional data onto an orthogonal coordinate system using linear transformations, extracting the principal components that best explain the data variance. The contribution rate of the first principal component represents the proportion of the most prevalent linearly correlated structures in the data; the larger the contribution rate of the first principal component, the stronger the linear correlation between different variables in the data. Therefore, the contribution rate of the first principal component indicates the linear coupling strength between the variables of each effect in the multi-effect evaporator group under the corresponding control cycle's operating state. The larger the contribution rate of the first principal component, the stronger the linear correlation and linear coupling effect between the thermodynamic parameters of each effect.

[0029] The contribution rate of the first principal component can only capture linear relationships. In order to further determine the nonlinear coupling relationship formed between each effect through thermodynamic cycle, the risk of oscillation of thermodynamic parameters of the multi-effect evaporator group is further assessed.

[0030] Within the same control cycle, based on the steam temperature sequence, pressure sequence, feed flow rate sequence, and liquid level sequence of the two effects, the mutual information entropy between the steam temperature, pressure, feed flow rate, and liquid level of the two effects is calculated respectively. The arithmetic mean of the four mutual information entropies is recorded as the corresponding thermodynamic parameter mutual information entropy of the two effects. The arithmetic mean of the normalized values ​​of the mutual information entropy of the thermodynamic parameters of all different effects in the multi-effect evaporator group is recorded as the comprehensive mutual information entropy of the same control cycle.

[0031] The calculation of mutual information entropy is a well-known technique and will not be elaborated further.

[0032] Mutual information entropy is used to quantify the degree of nonlinear dependence between two random variables. The larger the mutual information entropy, the stronger the nonlinear coupling between the two random variables. Therefore, the mutual information entropy of two-effect thermodynamic parameters represents the degree of nonlinear dependence between the two-effect thermodynamic parameters. The larger the mutual information entropy of thermodynamic parameters, the stronger the nonlinear coupling between the two-effect thermodynamic parameters, and the more uncontrollable the perturbation of the thermodynamic parameters.

[0033] The weighted sum of the contribution rate of the first principal component of the control cycle and the comprehensive mutual information entropy is denoted as the coupling risk index of the control cycle, where the weighted sum of the contribution rate of the first principal component of the control cycle and the comprehensive mutual information entropy is 1.

[0034] The weight of the contribution rate of the first principal component of the control cycle should be greater than or equal to 0.3 and less than or equal to 0.7. In this embodiment, the contribution rate of the first principal component of the control cycle and the weight of the comprehensive mutual information entropy are both 0.5.

[0035] The coupling risk index of the control cycle takes into account the influence of the linear and nonlinear relationships of thermodynamic parameters on the stability of the multi-effect evaporator group by weighted summation.

[0036] Thus, the coupling risk index of the control cycle is determined.

[0037] Step S003: Based on the differences between all thermodynamic parameter sequences of all different effects in the multi-effect evaporator group in the same control cycle, and the coupling risk index of the control cycle, calculate the control response sensitivity of the control cycle. The control response sensitivity is used to evaluate the synchronization degree of the responses of each effect in the multi-effect evaporator group.

[0038] Since the parameter coupling between each effect can generate a high risk of oscillation, even if the coupling strength between the thermodynamic parameters of each effect is low, the difference in the control action of each effect can lead to a decrease in the energy efficiency of the evaporator. Therefore, it is necessary to further quantify the interference of thermodynamic parameters on the control loop of each effect and the difference in control response.

[0039] Within the same control cycle, based on the steam temperature sequence, pressure sequence, feed flow rate sequence, and liquid level sequence of the two effects, the DTW distances for the steam temperature sequence, pressure sequence, feed flow rate sequence, and liquid level sequence of the two effects are calculated respectively. The arithmetic mean of the four DTW distances is recorded as the corresponding thermodynamic parameter DTW distance for the two effects. The coefficient of variation and arithmetic mean of the DTW distances of the thermodynamic parameters of all different effects in the multi-effect evaporator group are recorded as the thermodynamic parameter coefficient of variation and the average thermodynamic parameter DTW distance for the same control cycle, respectively.

[0040] The DTW distance, a thermodynamic parameter for two effects, is used to evaluate the degree of difference in the thermodynamic parameters between the two effects. The smaller the DTW distance, the smaller the difference in the thermodynamic parameters between the two effects, and the more synchronized the changes in the thermodynamic parameters of the two effects.

[0041] The coefficient of variation of thermodynamic parameters during the control period reflects the degree of dispersion of the thermodynamic parameter response modes among the effects in a multi-effect evaporator group. The smaller the coefficient of variation of thermodynamic parameters during the control period, the closer the DTW distance of the thermodynamic parameters of different effects in the multi-effect evaporator group is, that is, the more synchronous the response of each effect to the disturbance in time and the more consistent the change trajectory, and the more coordinated the overall response mode of the multi-effect evaporator group tends to be. The larger the coefficient of variation of thermodynamic parameters during the control period, the more significant the time lag or morphological difference in the response of each effect to the disturbance, and the greater the difficulty in the coordinated control of the multi-effect evaporator group.

[0042] The normalized value of the coefficient of variation of the thermodynamic parameters of the control cycle is denoted as the characteristic index value of the control cycle. The number 1 is used as the numerator. The product of the characteristic index value of the control cycle and the coupling risk index is denoted as the characteristic product. The normalized value of the average thermodynamic parameter DTW distance of the control cycle is calculated and the sum of the number 1 is calculated. The product of the sum and the characteristic product is used as the denominator. The normalized value of the fraction is denoted as the control response sensitivity of the control cycle.

[0043] In this embodiment, the sigmoid function is used to calculate the normalized value of the average thermodynamic parameter DTW distance. The maximum-minimum normalization method is used to calculate the normalized values ​​of the thermodynamic parameter variation coefficient and fraction. In the maximum-minimum normalization method, the sum of the difference between the maximum and minimum values ​​and a preset non-zero small constant is used as the denominator to calculate the normalized value. In this embodiment, the preset non-zero small constant is set to a value of... Among them, the sigmoid function and the maximum-minimum normalization method are well-known techniques and will not be described in detail here. As other implementation methods, implementers can use other methods of existing technology, such as the tanh function.

[0044] At this point, the control response sensitivity of the control cycle is obtained.

[0045] Step S004: Based on the control response sensitivity of the control cycle, the forgetting factor of the adaptive model predictive control algorithm is adaptively adjusted. Using the adaptive model predictive control algorithm, the control of the multi-effect evaporator group is realized based on the sequence of all thermodynamic parameters of all effects of the multi-effect evaporator group in the same control cycle.

[0046] During the operation of a multi-effect evaporator unit, changes in operating conditions such as equipment scaling and fluctuations in production load can cause a continuous deterioration in the control performance of the multi-effect evaporator unit. Therefore, an adaptive model predictive control algorithm is introduced as the core control algorithm to achieve real-time dynamic adjustment of the control strategy for the multi-effect evaporator unit.

[0047] The adaptive model predictive control algorithm can dynamically update the predictive model and solve for the optimal control sequence online within a single control cycle, based on real-time acquired thermodynamic parameters and state estimation results of the multi-effect evaporator group. Compared with traditional model predictive control algorithms, this algorithm has better adaptability to time-varying operating conditions and can correct model parameters in real time when the operating characteristics of the multi-effect evaporator group change, thus maintaining stable overall control performance. Specifically, using the temperature, pressure, liquid level, and feed flow rate sequences of each effect within the current control cycle as inputs to the adaptive model predictive control algorithm, it can output the optimal control output sequence for the next minute after the control cycle, i.e., the continuous control variables of the valve opening of each valve in the multi-effect evaporator group.

[0048] However, existing adaptive model predictive control algorithms only use a fixed-value forgetting factor to complete model iteration updates, which cannot sensitively detect dynamic changes in the response mode of multi-effect evaporator groups. When the synchronization of the operating responses of each evaporation unit is poor, local parameter disturbances can easily interfere with the model iteration process, further exacerbating the prediction model mismatch problem. The forgetting factor is used to limit the weight ratio of historical operating data in the algorithm model iteration process.

[0049] To overcome this technical deficiency, this application combines the control response sensitivity of the control cycle to dynamically and adaptively adjust the forgetting factor of the adaptive model predictive control algorithm.

[0050] Specifically, the adaptive calculation method for the forgetting factor in the control cycle of the adaptive model predictive control algorithm is as follows: the product of the difference between the number 1 and the control response sensitivity of the control cycle and a preset adjustment coefficient is recorded as the sensitivity product of the control cycle; the difference between the preset initial forgetting factor and the sensitivity product of the control cycle is taken as the characteristic forgetting factor of the control cycle. The characteristic forgetting factor of the control cycle is the adaptive value of the forgetting factor in the control cycle of the adaptive model predictive control algorithm.

[0051] It should be noted that when the feature forgetting factor of the control cycle is less than or equal to 0.9, the feature forgetting factor of the control cycle will be assigned a value of 0.9.

[0052] The adjustment coefficient should be greater than or equal to 0.05 and less than or equal to 0.1. In this embodiment, the adjustment coefficient is set to 0.05.

[0053] The closer the adaptive value of the forgetting factor is to 1, the slower the update and iteration speed, the better the adaptive model predictive control algorithm suppresses operating noise, but the lower its sensitivity to rapid changes in operating conditions of the multi-effect evaporator group.

[0054] The characteristic forgetting factor of the control cycle is used as the adaptive value of the forgetting factor of the adaptive model predictive control algorithm in the control cycle. The adaptive model predictive control algorithm is used to process all thermodynamic parameter sequences of all effects of the multi-effect evaporator group in the same control cycle to obtain the optimal control output sequence within 1 minute after the control cycle.

[0055] The optimal control output sequence contains the continuous control variables of the valve opening degree of each valve in the multi-effect evaporator group at each acquisition time within 1 minute after the control cycle.

[0056] Understandably, dynamically adjusting the update strategy of the adaptive model predictive control algorithm based on the operating status of the multi-effect evaporator group can enhance its adaptability to time-varying characteristics, thereby effectively suppressing parameter oscillations and improving the operating stability and control accuracy of the multi-effect evaporator group.

[0057] Thus, based on the coordinated optimization of thermodynamic parameters, the control of the multi-effect evaporator group is realized.

[0058] This application also proposes a multi-effect evaporator group based on thermodynamic parameter co-optimization, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps described above. Since the control method for the multi-effect evaporator group based on thermodynamic parameter co-optimization has been described in detail above, it will not be repeated here.

Claims

1. A control method for a multi-effect evaporator group based on coordinated optimization of thermodynamic parameters, characterized in that, The method includes the following steps: Thermodynamic parameters of each effect in the multi-effect evaporator group are collected at different collection times, and a sequence of thermodynamic parameters of each effect in each control cycle is established. The thermodynamic parameters include steam temperature, pressure, feed flow rate and liquid level. Based on the sequence of all thermodynamic parameters and the corresponding mutual information entropy of all different effects of the multi-effect evaporator group in the same control cycle, the coupling risk index of the control cycle is calculated. The coupling risk index is used to evaluate the impact of the linear and nonlinear relationships of thermodynamic parameters on the stability of the multi-effect evaporator group. Based on the differences between all thermodynamic parameter sequences of all different effects in the same control cycle of the multi-effect evaporator group, and the coupling risk index of the control cycle, the control response sensitivity of the control cycle is calculated. The control response sensitivity is used to evaluate the degree of synchronization of the responses of each effect in the multi-effect evaporator group. Based on the control response sensitivity of the control cycle, the forgetting factor of the adaptive model predictive control algorithm is adaptively adjusted. Using the adaptive model predictive control algorithm, the control of the multi-effect evaporator group is realized based on the sequence of all thermodynamic parameters of all effects of the multi-effect evaporator group in the same control cycle. The method for obtaining the coupling risk index of the control cycle is as follows: Based on the sequence of all thermodynamic parameters of all effects of the multi-effect evaporator group in the same control cycle, the contribution rate of the first principal component in the same control cycle is extracted; In the same control cycle, based on the steam temperature sequence, pressure sequence, feed flow rate sequence, and liquid level sequence of the two effects, the mutual information entropy between the steam temperature, pressure, feed flow rate, and liquid level of the two effects is calculated respectively, and the comprehensive mutual information entropy of the same control cycle is calculated. The weighted sum of the contribution rate of the first principal component of the control cycle and the comprehensive mutual information entropy is denoted as the coupling risk index of the control cycle, where the weighted sum of the contribution rate of the first principal component of the control cycle and the comprehensive mutual information entropy is 1. The method for adaptively adjusting the forgetting factor of the adaptive model predictive control algorithm based on the control response sensitivity of the control cycle includes the following specific methods: The product of the difference between the number 1 and the control response sensitivity of the control cycle and the preset adjustment coefficient is recorded as the sensitivity product of the control cycle. The difference between the preset initial forgetting factor and the sensitivity product of the control cycle is used as the adaptive value of the forgetting factor of the adaptive model predictive control algorithm in the control cycle.

2. The control method for a multi-effect evaporator group based on coordinated optimization of thermodynamic parameters according to claim 1, characterized in that, The method for obtaining the contribution rate of the first principal component of the control cycle is as follows: Based on the thermodynamic parameter sequences of all effects of the multi-effect evaporator group in the same control cycle, a thermodynamic parameter matrix for the same control cycle is established, and the contribution rate of the first principal component in the same control cycle is extracted using a principal component analysis algorithm.

3. The control method for a multi-effect evaporator group based on coordinated optimization of thermodynamic parameters according to claim 1, characterized in that, The method for obtaining the comprehensive mutual information entropy of the control cycle is as follows: The arithmetic mean of the four mutual information entropies is denoted as the corresponding thermodynamic parameter mutual information entropy of the two effects. The arithmetic mean of the normalized values ​​of the mutual information entropies of the thermodynamic parameters of all different effects of the multi-effect evaporator group is denoted as the comprehensive mutual information entropy of the same control cycle.

4. The control method for a multi-effect evaporator group based on coordinated optimization of thermodynamic parameters according to claim 1, characterized in that, The method for obtaining the control response sensitivity of the control cycle is as follows: In the same control cycle, the DTW distances of the steam temperature sequence, pressure sequence, feed flow rate sequence and liquid level sequence of the two effects are calculated respectively. The arithmetic mean of the four DTW distances is recorded as the corresponding thermodynamic parameter DTW distance of the two effects. The thermodynamic parameter variation coefficient and the average thermodynamic parameter DTW distance of the control cycle are determined. The normalized value of the coefficient of variation of the thermodynamic parameters of the control cycle is denoted as the characteristic index value of the control cycle. The number 1 is used as the numerator. The product of the characteristic index value of the control cycle and the coupling risk index is denoted as the characteristic product. The normalized value of the average thermodynamic parameter DTW distance of the control cycle is calculated and the sum of the number 1 is calculated. The product of the sum and the characteristic product is used as the denominator. The normalized value of the fraction is denoted as the control response sensitivity of the control cycle.

5. The control method for a multi-effect evaporator group based on coordinated optimization of thermodynamic parameters according to claim 4, characterized in that, The coefficient of variation of the thermodynamic parameters of the control cycle is: the coefficient of variation of the DTW distance of the thermodynamic parameters of all different effects of the multi-effect evaporator group.

6. The control method for a multi-effect evaporator group based on coordinated optimization of thermodynamic parameters according to claim 4, characterized in that, The average thermodynamic parameter DTW distance is the arithmetic mean of all different effects in the multi-effect evaporator group.

7. The control method for a multi-effect evaporator group based on coordinated optimization of thermodynamic parameters according to claim 1, characterized in that, The control method for the multi-effect evaporator group is as follows: based on the control sequence output by the adaptive model predictive control algorithm, the opening degree of each valve of the multi-effect evaporator group is controlled at each sampling time within one minute after the control cycle.

8. A multi-effect evaporator assembly based on thermodynamic parameter co-optimization, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the multi-effect evaporator group control method based on thermodynamic parameter collaborative optimization as described in any one of claims 1-7.

Images