An unattended intelligent control method and system for a heat exchange station

Abstract

The application discloses a kind of heat exchange station unattended intelligent control method and system, it is related to heat energy automatic control technical field, including collecting multi-source environmental parameter data, constructs environmental parameter sequence set, constructs heat load prediction model, by extracting local temperature trend feature and time series change relationship, output predicted heat load value;Actual heat load value and the residual error comparison of the predicted heat load value, construct residual error fluctuation measure sequence, and combined with preset threshold range multidimensional state discrimination, output operating status identification, and the frequency parameter of water pump and the opening parameter of water mixing valve are linked and adjusted, and call optimization control strategy set, generate updated water pump adjustment instruction and valve adjustment instruction.The application not only improves the operation reliability and energy utilization efficiency of heat exchange station, but also realizes unattended intelligent control, solves the technical problems of low prediction accuracy, inaccurate abnormal identification and poor control parameter coordination in traditional control method.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic control of thermal energy, in particular to a heat exchange station unattended intelligent control method and system. BACKGROUND

[0002] The current operation and management of heat exchange stations generally have high dependence on manpower, slow response to regulation and control, and low energy utilization efficiency. In the traditional mode, the operating personnel need to monitor the inlet and outlet temperatures and pressures of the plate heat exchanger and the water replenishment state in the system in real time. If a sudden situation such as abnormal pressure or water level occurs, a delayed response may cause a decrease in heating efficiency or even equipment failure. In addition, due to the inability to achieve unattended operation, the operation cost of the heat exchange station in some areas is high, the manual inspection intensity is large, the equipment state data is difficult to upload and remotely control in time, and it is unable to meet the urgent needs of efficient, energy-saving and safe intelligent heating systems. Therefore, how to realize the automatic and intelligent control of the heat exchange station to support unattended operation, and to realize real-time acquisition of key operation parameters, remote data interaction and state feedback, has become a key problem that needs to be solved at present.

[0003] CN103438503A discloses an intelligent control method and system for unattended operation of a heat exchange station. The output end of the programmable controller PLC of the intelligent control system is connected with two frequency converters to form a programmable controller PLC. The input end of the programmable controller PLC is connected with an outdoor temperature compensator, temperature and pressure sensors of the inlet and outlet of the high-temperature water on the primary side of the plate heat exchanger, temperature and pressure sensors of the inlet and outlet of the return water on the secondary side of the plate heat exchanger, and a liquid level sensor in the water replenishment tank. The corresponding output end is connected with an electric regulating valve on the high-temperature water on the primary side of the plate heat exchanger, an electromagnetic valve connected with the tap water pipeline, and an electromagnetic valve connected with the pipeline. The output end of the first frequency converter is connected with a circulating pump, and the output end of the second frequency converter is connected with a water replenishment pump. The system can realize automatic constant temperature and pressure heating, automatic water replenishment, abnormal alarm and power failure restart functions, has certain intelligent characteristics, and can send real-time field data to the control terminal for remote access and monitoring, thereby improving the management efficiency and operation reliability. Although the above-mentioned technology can realize a certain degree of automatic control and remote data transmission, the system structure is still mainly based on fixed logic PLC control, lacks dynamic adjustment and intelligent optimization capability, and when facing external environmental changes (such as sudden load change and sudden temperature difference fluctuation), the regulation and control response is lagged, and the adaptive operation strategy cannot be realized. At the same time, the scheme does not involve higher-level intelligent control modes such as edge computing, fault prediction analysis or multi-source data fusion, and still needs to rely on management personnel for strategy configuration and manual intervention in actual operation, and it is difficult to realize the true sense of "intelligent unattended operation", especially in large-scale central heating systems, it is difficult to meet the technical requirements of low energy consumption, high response and less maintenance. SUMMARY

[0004] In view of the problems of low energy efficiency, strong artificial dependence, insufficient remote monitoring and intelligent adjustment ability of the existing existing heat exchange station in the operation process, the present application is proposed.

[0005] Therefore, the problem to be solved by the present application is how to realize real-time acquisition and intelligent linkage control of key operating parameters of the heat exchange station, support remote management and fault adaptive processing, so as to realize efficient and safe unattended operation mode.

[0006] To solve the above technical problems, the present application provides the following technical solutions:

[0007] In a first aspect, the present application embodiment provides a kind of heat exchange station unattended intelligent control method, it includes,

[0008] Collecting multi-source environmental parameter data, constructing environmental parameter sequence set;

[0009] Based on the environmental parameter sequence set, a heat load prediction model is constructed, the local temperature trend feature and the time sequence change relationship are extracted, and the predicted heat load value is output;

[0010] Residual error comparison is carried out between actual heat load value and the predicted heat load value, residual error fluctuation degree sequence is constructed, and multi-dimensional state discrimination is carried out in combination with pre-set threshold range, and operation state identification is output;

[0011] Based on the operation state identification, linkage adjustment is carried out on water pump frequency parameter and water mixing valve opening degree parameter, and optimization control strategy set is called, and updated water pump adjustment instruction and valve adjustment instruction are generated.

[0012] As a preferred scheme of the heat exchange station unattended intelligent control method of the present application, wherein: based on the operation state identification, linkage adjustment is carried out on water pump frequency parameter and water mixing valve opening degree parameter, and optimization control strategy set is called, and updated water pump adjustment instruction and valve adjustment instruction are generated, including:

[0013] Current water pump frequency parameter and water mixing valve opening degree parameter are read from heat exchange station control system, and corresponding optimization control strategy set is selected according to operation state identification, wherein the optimization control strategy set includes normal condition strategy and abnormal condition strategy;

[0014] When operation state identification is normal state, water pump frequency adjustment interval and water mixing valve opening degree adjustment interval are divided according to the normal condition strategy, and current parameters are substituted into the normal condition strategy to calculate target adjustment amount;

[0015] When operation state identification is abnormal state, the abnormal condition strategy is started, adjustment direction is determined according to the fluctuation trend of residual error sequence, and the range of water pump frequency adjustment interval and water mixing valve opening degree adjustment interval is expanded.

[0016] generating a water pump adjustment instruction and a valve adjustment instruction according to the target adjustment amount, wherein the water pump adjustment instruction comprises a target frequency value; and the valve adjustment instruction comprises a target opening value;

[0017] sending the water pump adjustment instruction to a frequency converter and sending the valve adjustment instruction to an actuator, and recording an adjustment result in a history database.

[0018] As a preferred scheme of the heat exchange station unattended intelligent control method, the method for obtaining the operation state identifier comprises:

[0019] calculating an actual heat load value according to a primary supply-return water temperature difference, a secondary supply-return water temperature difference and a heat exchange station flow at the current time, and recording a difference between the actual heat load value and the predicted heat load value as a residual value;

[0020] arranging the residual value in time sequence to form a residual sequence, calculating a mean value, a standard deviation and a coefficient of variation of the residual sequence to obtain residual statistical characteristic values;

[0021] establishing a first threshold interval, a second threshold interval and a third threshold interval according to normal operation data calibrated in advance;

[0022] comparing the residual statistical characteristic values with the first threshold interval, the second threshold interval and the third threshold interval to mark abnormal points;

[0023] counting a number and a distribution of the abnormal points within a preset time, and marking a current heat exchange station operation state as an abnormal state when the number of the abnormal points exceeds a preset number threshold or continuous abnormal points occur.

[0024] As a preferred scheme of the heat exchange station unattended intelligent control method, the comparison of the residual statistical characteristic values with the first threshold interval, the second threshold interval and the third threshold interval comprises:

[0025] if the mean value of the residual sequence is located in the first threshold interval and the standard deviation is less than a lower limit of the second threshold interval, determining that the abnormal point is a regulation-insensitive abnormal point;

[0026] if the mean value of the residual sequence exceeds an upper limit of the first threshold interval and the coefficient of variation exceeds an upper limit of the third threshold interval, determining that the abnormal point is an over-regulation abnormal point;

[0027] if the mean value of the residual sequence is located in the first threshold interval, but the standard deviation exceeds an upper limit of the second threshold interval and the coefficient of variation is located in the third threshold interval, determining that the abnormal point is a steady-state fluctuation abnormal point;

[0028] If the change rate of the residual sequence mean exceeds the preset change rate and the coefficient of variation is lower than the lower limit of the third threshold interval, the mutation abnormal point is determined.

[0029] As a preferred scheme of the heat exchange station unattended intelligent control method, the method for obtaining the predicted heat load value comprises the following steps:

[0030] The environment parameter sequence set is divided into a plurality of training sample subsets according to a preset time window.

[0031] The primary supply and return water temperature and the secondary supply and return water temperature in the training sample subset are subjected to difference operation to obtain a primary temperature difference and a secondary temperature difference, and the primary temperature difference and the secondary temperature difference and the outdoor temperature at the corresponding moment are combined to form a temperature feature vector.

[0032] The temperature feature vector is associated and paired with the water pump frequency and the heat network pressure at the corresponding moment to construct an input feature matrix, and the historical heat load data at the corresponding moment is taken as an output label.

[0033] The input feature matrix and the output label are trained by using a support vector regression algorithm to obtain kernel function parameters and slack variables of a heat load prediction model.

[0034] A local temperature trend correction factor is introduced, the preliminary predicted value is weighted and corrected according to the outdoor temperature change rate, and a predicted heat load value is generated.

[0035] As a preferred scheme of the heat exchange station unattended intelligent control method, the method for obtaining the predicted heat load value comprises the following steps:

[0036] The input feature matrix is subjected to normalization processing, and the environment parameter sequence set is mapped to the [0, 1] interval to obtain a standardized feature matrix.

[0037] A support vector regression model is constructed by using a radial basis kernel function, a kernel function is defined, and a heat load prediction model is formed.

[0038] Based on the standardized feature matrix and the output label, an ε-insensitive loss function is used to establish a regression target function and a constraint condition.

[0039] The regression target function is solved by using a sequential minimal optimization algorithm to obtain an optimal kernel width parameter and a penalty factor, and an optimal hyperplane parameter of the heat load prediction model is determined.

[0040] As a preferred scheme of the heat exchange station unattended intelligent control method, the environment parameter sequence set comprises a primary supply and return water temperature, a secondary supply and return water temperature, a water pump frequency, an outdoor temperature, a heat network pressure, and historical heat load data.

[0041] In a second aspect, an embodiment of the present application provides an unattended intelligent control system for a heat exchange station, comprising:

[0042] An environmental parameter acquisition and sequence construction module is configured to acquire multi-source environmental parameter data and construct an environmental parameter sequence set.

[0043] A heat load prediction model construction module is configured to construct a heat load prediction model based on the environmental parameter sequence set, extract local temperature trend features and time sequence change relations, and output a predicted heat load value.

[0044] A residual error analysis and state discrimination module is configured to compare an actual heat load value with the predicted heat load value, construct a residual error fluctuation metric sequence, perform multi-dimensional state discrimination in combination with a preset threshold range, and output an operating state identifier.

[0045] An optimized control strategy execution module is configured to perform linkage adjustment on a water pump frequency parameter and a water mixing valve opening degree parameter based on the operating state identifier, and call an optimized control strategy set to generate updated water pump adjustment instructions and valve adjustment instructions.

[0046] In a third aspect, an embodiment of the present application provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program instructions are executed by the processor, the steps of the unattended intelligent control method for a heat exchange station according to the first aspect of the present application are implemented.

[0047] In a fourth aspect, an embodiment of the present application provides a computer readable storage medium having a computer program stored thereon, wherein when the computer program instructions are executed by a processor, the steps of the unattended intelligent control method for a heat exchange station according to the first aspect of the present application are implemented.

[0048] The present application has the following beneficial effects: By acquiring multi-source environmental parameter data and constructing an environmental parameter sequence set, the system's comprehensive perception ability for the operating environment of the heat exchange station is ensured; by using a heat load prediction model constructed based on a support vector regression algorithm and a local temperature trend correction factor, accurate prediction of heat load changes is achieved; based on residual error analysis and multi-dimensional state discrimination mechanisms for actual heat load and predicted heat load, various abnormal states such as non-sensitive adjustment, over-adjustment, steady-state fluctuation, and mutation can be accurately identified, and fault diagnosis capability is improved; through linkage adjustment of a water pump frequency parameter and a water mixing valve opening degree parameter and an optimized control strategy set for different operating states, efficient and stable operation of the heat exchange station is achieved; the present application not only significantly improves the operating reliability and energy utilization efficiency of the heat exchange station, but also realizes true unattended intelligent control, solves the technical problems of low prediction accuracy, inaccurate abnormality identification, and poor control parameter coordination in traditional control methods, and provides an effective solution for the intelligent upgrading of heating systems. Attached Figure Description

[0049] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0050] Figure 1 This is a flowchart of the unattended intelligent control method for the heat exchange station in Example 1.

[0051] Figure 2 This is a system structure diagram of the unattended intelligent control system for the heat exchange station in Example 2. Detailed Implementation

[0052] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0053] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0054] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0055] Example 1

[0056] Reference Figure 1 This is the first embodiment of the present invention, which provides an unattended intelligent control method for a heat exchange station, including:

[0057] S1: Collect multi-source environmental parameter data and construct an environmental parameter sequence set.

[0058] S1.1: The primary supply water temperature and primary return water temperature are collected by the first temperature sensor installed in the primary pipeline of the heat exchange station.

[0059] S1.2: At the same time, the secondary supply water temperature and the secondary return water temperature are collected by a second temperature sensor installed in the secondary pipeline of the heat exchange station.

[0060] S1.3: Collecting the primary pipe network heat network pressure through the first pressure sensor arranged at the heat exchange station and collecting the water pump frequency data through the rotation speed sensor arranged at the water pump.

[0061] S1.4: Collecting the outdoor temperature through the meteorological sensor arranged outside the heat exchange station and reading the historical heat load data from the database.

[0062] S1.5: Arranging the primary water supply temperature, the primary return water temperature, the secondary water supply temperature, the secondary return water temperature, the heat network pressure, the water pump frequency data, the outdoor temperature and the historical heat load data in the time stamp order to form an environmental parameter sequence set.

[0063] S1.6: Storing the environmental parameter sequence set into the data buffer area and eliminating the abnormal data in the environmental parameter sequence set to obtain a corrected environmental parameter sequence set.

[0064] It should be noted that the environmental parameter sequence set includes the primary supply and return water temperature, the secondary supply and return water temperature, the water pump frequency, the outdoor temperature, the heat network pressure and the historical heat load data.

[0065] S2: Based on the environmental parameter sequence set, a heat load prediction model is constructed, the local temperature trend feature and the time sequence change relationship are extracted, and the predicted heat load value is output.

[0066] S2.1: Dividing the environmental parameter sequence set into a plurality of training sample subsets according to a preset time window.

[0067] S2.2: Carrying out difference operation on the primary supply and return water temperature and the secondary supply and return water temperature in the training sample subset to obtain the primary temperature difference and the secondary temperature difference, and combining the primary temperature difference and the secondary temperature difference with the outdoor temperature at the corresponding moment to form a temperature feature vector.

[0068] S2.3: Associating and pairing the temperature feature vector with the water pump frequency and the heat network pressure at the corresponding moment to construct an input feature matrix, and taking the historical heat load data at the corresponding moment as an output label.

[0069] S2.4: Training the input feature matrix and the output label by using a support vector regression algorithm to obtain the kernel function parameter and the slack variable of the heat load prediction model.

[0070] Specifically, it includes:

[0071] S2.4.1: Normalizing the input feature matrix to map the environmental parameter sequence set to the interval [0, 1] to obtain a standardized feature matrix;

[0072] S2.4.2: Constructing a support vector regression model by using a radial basis kernel function, defining a kernel function to form a heat load prediction model;

[0073] The specific formula of the heat load prediction model is preferably as follows:

[0074]

[0075] wherein, is the predicted heat load value, and is the Lagrange multiplier, is the temperature trend correction factor, and b is the bias term, is the i-th training sample feature vector, is the current input feature vector, is the number of support vectors, is the local temperature trend correction factor, is the improved radial basis kernel function, and the specific formula is as follows:

[0076]

[0077] wherein, σ is the kernel function width parameter, is the j-th feature component of the i-th sample, is the j-th feature component of the current input, is the dynamic weight of the j-th feature, and the specific formula is as follows:

[0078]

[0079] wherein, is the dynamic weight of the j-th feature, is the feature weight adjustment parameter, and is 0.5, is the standard deviation of the j-th feature, and m is the feature dimension.

[0080] It should be noted that, the value range of is [0, ], wherein is the rated heat load of the heat exchange station; the value range of is [0, 1], indicating the feature weight.

[0081] S2.4.3: Based on the standardized feature matrix and the output label, an ε-insensitive loss function is used to establish a regression objective function and a constraint condition;

[0082] S2.4.4: The regression objective function is solved by a sequential minimal optimization algorithm to obtain the optimal kernel width parameter and the penalty factor, and to determine the optimal hyperplane parameter of the heat load prediction model.

[0083] S2.5: A local temperature trend correction factor is introduced, and the preliminary predicted value is weighted and corrected according to the outdoor temperature change rate to generate a predicted heat load value. ​​​

[0084] The specific formula of the local temperature trend correction factor is preferably as follows:

[0085] ;

[0086] wherein, T is the outdoor temperature, T is the reference temperature value, is the temperature change rate influence coefficient, taken as 3600s, is the primary supply and return water temperature difference, is the secondary supply and return water temperature difference.

[0087] In an optional embodiment, the latest collected environmental parameter sequence and the historical data are recombined at a preset ratio to dynamically adjust the time sequence correlation weight of the input feature matrix; the adjusted input feature matrix is input into the heat load prediction model, the high-dimensional space similarity is calculated through a radial basis kernel function, and the preliminary prediction value is output in combination with the optimal hyperplane parameter.

[0088] S3: The actual heat load value and the predicted heat load value are compared for residual error, a residual error fluctuation degree sequence is constructed, and multi-dimensional state discrimination is performed in combination with a preset threshold range to output a running state identifier.

[0089] S3.1: The actual heat load value is calculated according to the primary supply and return water temperature difference, the secondary supply and return water temperature difference, and the heat exchange station flow at the current time, and the difference between the actual heat load value and the predicted heat load value is recorded as a residual error value.

[0090] The specific formula of the residual error value is preferably as follows:

[0091] ;

[0092] wherein, is the residual error value at time t, is the actual heat load value at time t, is the predicted heat load value at time t, is the reference heat load value, taken as 20% of the rated heat load, is the flow fluctuation correction factor.

[0093] S3.2: The residual error values are arranged in time sequence to form a residual error sequence, the mean, the standard deviation, and the coefficient of variation of the residual error sequence are calculated to obtain residual error statistical characteristic values.

[0094] S3.3: The first threshold interval, the second threshold interval, and the third threshold interval are established according to the pre-labeled normal running data.

[0095] It should be noted that the first threshold interval corresponds to the mean value range of the residual error sequence, the second threshold interval corresponds to the standard deviation range, and the third threshold interval corresponds to the coefficient of variation range.

[0096] For example, the threshold criteria established according to the historical operation data of the system are as follows: the first threshold interval (mean value range) is set as [-20kW, 20kW] for evaluating the overall level of the prediction deviation; the second threshold interval (standard deviation range) is [5kW, 15kW] for measuring the fluctuation amplitude of the data; and the third threshold interval (coefficient of variation range) is [0.1, 0.3] for judging the stability of the fluctuation.

[0097] S3.4: Compare the residual statistical characteristic value with the first threshold interval, the second threshold interval and the third threshold interval, and mark the abnormal points.

[0098] In an optional embodiment, if the mean value of the residual sequence is located in the first threshold interval and the standard deviation is less than the lower limit of the second threshold interval, it is determined as a regulation-insensitive abnormal point; if the mean value of the residual sequence exceeds the upper limit of the first threshold interval and the coefficient of variation exceeds the upper limit of the third threshold interval, it is determined as an over-regulation abnormal point; if the mean value of the residual sequence is located in the first threshold interval, but the standard deviation exceeds the upper limit of the second threshold interval and the coefficient of variation is located in the third threshold interval, it is determined as a steady-state fluctuation abnormal point; and if the change rate of the mean value of the residual sequence exceeds the preset change rate and the coefficient of variation is lower than the lower limit of the third threshold interval, it is determined as a sudden change abnormal point.

[0099] Specifically, in the operation monitoring process of the heat exchange station, the statistical characteristic analysis of the residual sequence is a key link for distinguishing the system state; with 1 hour as the sampling period, the residual values (unit: kW) of the continuous 6 sampling points are: [-15, -12, -10, 5, 8, 20] in turn. The following statistical characteristics can be obtained by calculation: the arithmetic mean of the residual sequence is -0.67kW, which reflects that the prediction value is overall slightly higher than the actual value in this period; the standard deviation calculation result is 13.2kW, indicating that the data dispersion degree is large; and the coefficient of variation is about 0.197 after standardization, which reflects the relative fluctuation level of the data.

[0100] For example, if the residual mean is -5 kW (within the first threshold interval) and the standard deviation is 4 kW (lower than the lower limit of the second threshold by 5 kW), it is determined that the adjustment is insensitive to abnormal points, indicating that the system response is slow. If the residual mean suddenly increases to 25 kW (exceeding the upper limit of the first threshold by 20 kW) and the coefficient of variation reaches 0.35 (exceeding the upper limit of the third threshold by 0.3), it is determined that the adjustment is over-adjusted, indicating that the control command overshoots and the adjustment amplitude needs to be suppressed. If the residual mean remains at 10 kW (within the first threshold interval), but the standard deviation rises to 18 kW (exceeding the upper limit of the second threshold by 15 kW) and the coefficient of variation is 0.2 (within the third threshold interval), it is determined that the adjustment is a steady-state fluctuation abnormal point, indicating that the system has periodic fluctuations. If the residual mean of the adjacent two sampling points jumps from -10 kW to 15 kW (with a change rate of 25 kW / h, exceeding the preset change rate of 20 kW / h) and the coefficient of variation drops to 0.08 (lower than the lower limit of the third threshold of 0.1), it is determined that the adjustment is a sudden abnormal point, indicating a sudden disturbance (such as valve failure).

[0101] Actual application scenario: On a certain day, the residual sequence mean is 18 kW (normal), the standard deviation is 4 kW (normal), but the coefficient of variation suddenly drops to 0.05 (abnormally low), so combined with the change rate exceeding the limit, the system is marked as a sudden abnormal point, triggering an emergency check to see if the water pump is jammed.

[0102] S3.5: Count the number and distribution of abnormal points within a preset time, and when the number of abnormal points exceeds a preset number threshold or consecutive abnormal points appear, mark the current heat exchange station running state as an abnormal state.

[0103] In an optional implementation, when one of the following conditions occurs within 24 hours, the running state is identified as an abnormal state: adjustment insensitive abnormal points occur in 3 consecutive sampling periods; over-adjustment abnormal points occur in 2 consecutive sampling periods; the cumulative number of steady-state fluctuation abnormal points within any 4 hours exceeds 35% of the total number of sampling points; after the adjustment divergence abnormal point appears, the residual mean of the next sampling period does not return to the first threshold interval; after the sudden abnormal point appears, the coefficient of variation of the next 2 sampling periods is continuously lower than the lower limit of the third threshold interval.

[0104] In an optional implementation, when the following conditions are met within 24 hours, the running state is identified as a normal state: the residual sequence mean is always within the first threshold interval; the standard deviation fluctuation range does not exceed 80% of the second threshold interval; the maximum value of the coefficient of variation does not exceed 90% of the upper limit of the third threshold interval; the single duration of any type of abnormal point does not exceed 2 sampling periods; the total number of abnormal points within 24 hours accounts for less than 20% of the total number of sampling points.

[0105] S4: Based on the operating status identifier, the pump frequency parameter and mixing valve opening parameter are adjusted in a coordinated manner, and the optimized control strategy set is called to generate updated pump adjustment instructions and valve adjustment instructions.

[0106] S4.1: Read the current pump frequency parameters and mixing valve opening parameters from the heat exchange station control system, and select the corresponding optimized control strategy set according to the operating status identifier. The optimized control strategy set includes normal operating condition strategy and abnormal operating condition strategy.

[0107] S4.2: When the operating status is marked as normal, the pump frequency adjustment range and mixing valve opening adjustment range are divided according to the normal operating condition strategy, and the current parameters are substituted into the normal operating condition strategy to calculate the target adjustment amount.

[0108] Preferably, the optimized control strategy set is pre-established based on historical operating data, typical operating condition response patterns and energy consumption constraints, and is divided into normal operating condition strategies and abnormal operating condition strategies.

[0109] Specifically, the normal operating condition strategies mainly include steady-state control strategy, smooth transition strategy, and energy efficiency optimization strategy. The steady-state control strategy limits the pump frequency adjustment range to 40%~80% of the rated frequency and the mixing valve opening adjustment range to 20%~80%, while limiting the single adjustment step size (frequency ≤ 2Hz, opening ≤ 5%) and setting a minimum response time of no less than 300 seconds. The smooth transition strategy adopts a gradient adjustment method, maintaining a 2:1 ratio between the coordinated changes in pump frequency and valve opening. When the residual is less than 5%, a dead zone is set to prevent further adjustment. The energy efficiency optimization strategy prioritizes adjusting the mixing valve opening while meeting heating requirements, maintaining the lowest feasible pump frequency, and avoiding frequent small adjustments to achieve stable system operation and energy saving.

[0110] Furthermore, abnormal operating condition strategies include rapid response strategies, anomaly handling strategies, and safety protection strategies; rapid response strategies: by expanding the water pump frequency adjustment range to The mixing valve opening range is adjusted to [10%, 90%], while increasing the single adjustment step size (frequency ≤ 5Hz, opening ≤ 10%) and shortening the response time to no less than 120 seconds; Abnormal handling strategy: When the adjustment is insensitive, increase the adjustment step size and shorten the response time; when over-adjustment occurs, use reverse compensation adjustment and reduce the adjustment step size; when there is steady-state fluctuation, increase the adjustment dead zone and extend the response time; when there is a sudden abnormality, start the emergency adjustment mode; Safety protection strategy: Set hard constraints on the upper and lower limits of adjustment, introduce an anti-oscillation protection mechanism, and establish an adjustment divergence judgment and backoff mechanism to ensure the safe and reliable operation of the system under abnormal conditions.

[0111] S4.3: When the running state is identified as an abnormal state, then start the abnormal condition strategy, determine the adjustment direction according to the fluctuation trend of the residual sequence, and expand the range of the water pump frequency adjustment interval and the water valve opening adjustment interval.

[0112] S4.4: Generate a water pump adjustment instruction and a valve adjustment instruction according to the target adjustment amount, wherein the water pump adjustment instruction includes a target frequency value; the valve adjustment instruction includes a target opening value.

[0113] Specifically includes:

[0114] S4.4.1: Input the target adjustment amount into the adjustment amount conversion formula to calculate the water pump target frequency adjustment value and the valve target opening adjustment value;

[0115] Preferably, the specific formula of the water pump target frequency adjustment value and the valve target opening adjustment value is as follows:

[0116]

[0117] ;

[0118] Wherein, is the water pump target frequency adjustment value, is the current reference frequency value, is the frequency adjustment coefficient, taking a value of 0.1-0.5, is the heat load deviation amplification coefficient, is the heat load residual value, is the temperature change rate influence coefficient, is the outdoor temperature, is the valve target opening adjustment value, is the current reference opening value, is the opening adjustment coefficient, taking a value of 0.1-0.3, is the opening adjustment sensitivity coefficient.

[0119] S4.4.2: Read the current water pump frequency and valve opening value, superimpose the water pump target frequency adjustment value and the current water pump frequency value to obtain the water pump target frequency value, and superimpose the valve target opening adjustment value and the current valve opening value to obtain the valve target opening value;

[0120] In an optional embodiment, a frequency limit condition is applied to the water pump target frequency value: if the water pump target frequency value is less than the minimum frequency value, the water pump target frequency value is set to the minimum frequency value; if the water pump target frequency value is greater than the maximum frequency value, the water pump target frequency value is set to the maximum frequency value.

[0121] In an optional embodiment, an opening degree limit condition is applied to the valve target opening degree value: if the valve target opening degree value is less than the minimum opening degree value, the valve target opening degree value is set to the minimum opening degree value; if the valve target opening degree value is greater than the maximum opening degree value, the valve target opening degree value is set to the maximum opening degree value.

[0122] S4.4.5: generating a water pump adjustment instruction according to the water pump target frequency value, wherein the water pump adjustment instruction comprises a device type identifier, the water pump target frequency value, an adjustment timestamp, and an adjustment direction identifier;

[0123] S4.4.6: generating a valve adjustment instruction according to the valve target opening degree value, wherein the valve adjustment instruction comprises a device type identifier, the valve target opening degree value, an adjustment timestamp, and an adjustment direction identifier.

[0124] S4.5: sending the water pump adjustment instruction to a frequency converter and sending the valve adjustment instruction to an actuator, and recording an adjustment result in a historical database.

[0125] To sum up, the present application ensures the comprehensive sensing ability of the system to the operation environment of the heat exchange station by collecting multi-source environmental parameter data and constructing an environmental parameter sequence set; realizes the accurate prediction of the heat load change through the heat load prediction model constructed by the support vector regression algorithm and the local temperature trend correction factor; can accurately identify various abnormal states such as non-sensitive adjustment, over-adjustment, steady-state fluctuation, and mutation based on the residual analysis of the actual heat load and the predicted heat load and the multi-dimensional state discrimination mechanism, thereby improving the fault diagnosis capability; realizes the efficient and stable operation of the heat exchange station through the linkage adjustment of the water pump frequency parameter and the water mixing valve opening degree parameter and the optimization control strategy set for different operation states; the present application not only significantly improves the operation reliability and energy utilization efficiency of the heat exchange station, but also realizes the truly unattended intelligent control, solves the technical problems of low prediction accuracy, inaccurate abnormal identification, and poor control parameter coordination in the traditional control method, and provides an effective solution for the intelligent upgrading of the heating system.

[0126] Embodiment 2

[0127] With reference to Figure 2 For the second embodiment of the present application, the embodiment provides an unattended intelligent control system for a heat exchange station, comprising:

[0128] an environmental parameter acquisition and sequence construction module, configured to acquire multi-source environmental parameter data and construct an environmental parameter sequence set;

[0129] a heat load prediction model construction module, configured to construct a heat load prediction model based on the environmental parameter sequence set, extract local temperature trend features and time sequence change relations, and output a predicted heat load value;

[0130] The residual error analysis and state identification module is configured to compare the actual heat load value and the predicted heat load value, construct a residual error fluctuation sequence, and perform multi-dimensional state identification in combination with a preset threshold range, and output an operation state identifier.

[0131] The optimized control strategy execution module is configured to perform linkage adjustment on the water pump frequency parameter and the water mixing valve opening degree parameter based on the operation state identifier, and call the optimized control strategy set to generate updated water pump adjustment instructions and valve adjustment instructions.

[0132] It should be noted that the technical scheme of the heat exchange station unattended intelligent control system is the same as the technical scheme of the heat exchange station unattended intelligent control method, and the details of the technical scheme of the heat exchange station unattended intelligent control system in the embodiment are not described in detail, and can be referred to the description of the technical scheme of the heat exchange station unattended intelligent control.

[0133] The above-mentioned unit modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory in the computer device in software form, so as to call and execute the operations corresponding to the above-mentioned modules by the processor.

[0134] The embodiment also provides an electronic device, which comprises a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is configured to perform wired or wireless communication with an external terminal. The wireless communication can be realized through WIFI, an operator network, NFC (near field communication) or other technologies. The computer program is executed by the processor to realize a multi-task edge computing resource scheduling method. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device, or an external keyboard, touchpad or mouse, etc.

[0135] The embodiment also provides a computer readable storage medium having a computer program stored thereon, which is executed by the processor to realize the method proposed in the above-mentioned embodiment.

[0136] The storage medium proposed in the embodiment belongs to the same inventive concept as the method proposed in the above-mentioned embodiment, and the technical details not described in detail in the embodiment can be referred to the above-mentioned embodiment, and the embodiment has the same beneficial effects as the above-mentioned embodiment.

[0137] Through the above description of the embodiments, those skilled in the art can clearly understand that the present application can be realized by means of software and necessary universal hardware, and of course can also be realized by hardware, but in many cases the former is a better embodiment. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product, and the computer software product can be stored in a computer readable storage medium, such as a floppy disk, a read-only memory (ROM), a random access memory (RAM), a FLASH, a hard disk, or an optical disc, etc., and includes a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods of the embodiments of the present application.

[0138] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application but not limit the present application. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced equivalently without departing from the spirit and scope of the present application, and all of them should be covered in the scope of the claims of the present application.

[0139] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can adopt a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer usable storage media (including but not limited to a disk storage, a CD-ROM, an optical storage, etc.) containing computer usable program codes. The solutions in the embodiments of the present application can be implemented in various computer languages.

[0140] The present application is described with reference to flowcharts and / or block diagrams according to the methods, devices (systems), and computer program products of the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be realized by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to produce a machine, so that the instructions executed by the computer or other programmable data processing devices produce a device for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 The device for implementing the functions specified in one flow or multiple flows and / or blocks. Figure 1 The device for implementing the functions specified in one flow or multiple flows and / or blocks.

[0141] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the Figure 1 function specified in the flow or flows and / or blocks Figure 1 of the block or blocks.

[0142] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions that are executed on the computer or other programmable apparatus provide steps for implementing the Figure 1 function specified in the flow or flows and / or blocks Figure 1 of the block or blocks.

[0143] While the preferred embodiments of the application have been described, additional variations and modifications can be made to the preferred embodiments by those of skill in the art once they have the benefit of the present disclosure. Therefore, the appended claims are intended to encompass within their scope all possible variations and modifications of the preferred embodiments.

[0144] It is apparent that a person skilled in the art can make a variety of changes and modifications to the application without departing from the spirit and scope of the application. Thus, if these modifications and variations of the application fall within the scope of the claims and their equivalents, it is intended to include them within the scope of the application.

Claims

1. A method for unmanned intelligent control of a heat exchange station, characterized in that: include, Collect multi-source environmental parameter data and construct an environmental parameter sequence set; Based on the set of environmental parameter sequences, a heat load prediction model is constructed, and the predicted heat load value is output by extracting the relationship between local temperature trend features and time series changes. The actual heat load value and the predicted heat load value are compared with the residual to construct a residual fluctuation measurement sequence. Then, multi-dimensional state discrimination is performed in combination with a preset threshold range, and the operating status identifier is output. Based on the operating status identifier, the pump frequency parameter and mixing valve opening parameter are adjusted in a coordinated manner, and the optimized control strategy set is invoked to generate updated pump adjustment instructions and valve adjustment instructions. The current pump frequency parameters and mixing valve opening parameters are read from the heat exchange station control system, and the corresponding optimized control strategy set is selected according to the operating status identifier. The optimized control strategy set includes normal operating condition strategy and abnormal operating condition strategy. When the operating status is marked as normal, the pump frequency adjustment range and the mixing valve opening adjustment range are divided according to the normal operating condition strategy, and the current parameters are substituted into the normal operating condition strategy to calculate the target adjustment amount. When the operating status is marked as abnormal, the abnormal operating condition strategy is activated, and the adjustment direction is determined according to the fluctuation trend of the residual sequence. At the same time, the range of the pump frequency adjustment range and the mixing valve opening adjustment range are expanded. Based on the target adjustment amount, a water pump adjustment command and a valve adjustment command are generated, wherein the water pump adjustment command includes a target frequency value; and the valve adjustment command includes a target opening value. The water pump adjustment command is sent to the frequency converter, and the valve adjustment command is sent to the actuator. The adjustment result is recorded in the historical database. The method for obtaining the running status identifier is as follows: The actual heat load value is calculated based on the current temperature difference between the primary and secondary supply and return waters and the flow rate of the heat exchange station, and the difference between the actual heat load value and the predicted heat load value is recorded as the residual value. The residual values ​​are arranged in chronological order to form a residual sequence. The mean, standard deviation, and coefficient of variation of the residual sequence are calculated to obtain the residual statistical characteristic values. Establish a first threshold interval, a second threshold interval, and a third threshold interval based on pre-calibrated normal operation data; The residual statistical feature values ​​are compared with the first threshold interval, the second threshold interval, and the third threshold interval to mark outliers; The number and distribution of abnormal points within a preset time period are statistically analyzed. When the number of abnormal points exceeds the preset threshold or consecutive abnormal points occur, the current operating status of the heat exchange station is marked as abnormal. The method for constructing the heat load prediction model is as follows: The input feature matrix is ​​normalized to map the set of environmental parameter sequences to the [0,1] interval, thus obtaining the standardized feature matrix; A support vector regression model is constructed using a radial basis function kernel function, and the kernel function is defined to form a heat load prediction model. Based on the standardized feature matrix and output labels, an ε-insensitive loss function is used to establish the regression objective function and constraints. The regression objective function is solved by the sequential minimum optimization algorithm to obtain the optimal kernel width parameter and penalty factor, and the optimal hyperplane parameters of the heat load prediction model are determined. If the mean of the residual sequence is within the first threshold interval and the standard deviation is less than the lower limit of the second threshold interval, it is determined to be an adjustment-insensitive outlier. If the mean of the residual sequence exceeds the upper limit of the first threshold interval and the coefficient of variation exceeds the upper limit of the third threshold interval, it is determined to be an over-adjusted outlier. If the mean of the residual sequence is within the first threshold interval, but the standard deviation exceeds the upper limit of the second threshold interval and the coefficient of variation is within the third threshold interval, then it is determined to be a steady-state fluctuation anomaly. If the rate of change of the mean of the residual sequence exceeds the preset rate of change and the coefficient of variation is lower than the lower limit of the third threshold interval, it is determined to be a mutation anomaly.

2. The unmanned intelligent control method for heat exchange stations as described in claim 1, characterized in that: The method for obtaining the predicted heat load value is as follows: The set of environmental parameter sequences is divided into multiple training sample subsets according to a preset time window; The primary and secondary supply and return water temperatures in the training sample subset are used to perform a difference calculation to obtain the primary temperature difference and the secondary temperature difference. The primary temperature difference and the secondary temperature difference are then combined with the outdoor temperature at the corresponding time to form a temperature feature vector. The temperature feature vector is associated with the pump frequency and heating network pressure at the corresponding time to construct the input feature matrix, and the historical heat load data at the corresponding time is used as the output label. The input feature matrix and the output label are trained using the support vector regression algorithm to obtain the kernel function parameters and slack variables of the heat load prediction model; A local temperature trend correction factor is introduced to weight and correct the preliminary forecast value based on the outdoor temperature change rate, thereby generating the predicted heat load value.

3. The unmanned intelligent control method for heat exchange stations as described in claim 1, characterized in that: The set of environmental parameter sequences includes primary supply and return water temperatures, secondary supply and return water temperatures, water pump frequency, outdoor temperature, heating network pressure, and historical heat load data.

4. An unattended intelligent control system for a heat exchange station, based on the unattended intelligent control method for a heat exchange station as described in any one of claims 1 to 3, characterized in that: include, The environmental parameter acquisition and sequence construction module is used to collect multi-source environmental parameter data and construct an environmental parameter sequence set. The heat load prediction model construction module constructs a heat load prediction model based on the set of environmental parameter sequences, and outputs the predicted heat load value by extracting the relationship between local temperature trend features and time series changes. The residual analysis and state discrimination module is used to compare the actual heat load value and the predicted heat load value with residuals, construct a residual fluctuation measurement sequence, and perform multi-dimensional state discrimination in combination with a preset threshold range, and output the operating state identifier. The optimized control strategy execution module, based on the operating status identifier, performs coordinated adjustment of the pump frequency parameters and mixing valve opening parameters, and calls the optimized control strategy set to generate updated pump adjustment instructions and valve adjustment instructions.

5. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the unattended intelligent control method for heat exchange stations as described in any one of claims 1 to 3.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the unattended intelligent control method for heat exchange stations as described in any one of claims 1 to 3.

Images