Water pump dynamic model calibration and efficiency optimization method

Abstract

The invention relates to the technical field of water pump energy efficiency optimization, and discloses a water pump dynamic model calibration and efficiency optimization method. The method comprises the following steps: constructing a water pump dynamic characteristic model comprising a plurality of operation parameter nodes, an association relationship among the parameter nodes and characteristic information of each parameter node, and updating the model based on real-time operation data such as flow change, pressure fluctuation, rotating speed feedback and environment temperature; key influence parameters, low-efficiency parameter combinations and deviation nodes in the model are identified through a parameter sensitivity analysis algorithm; generating an initial calibration strategy set comprising a parameter compensation scheme, an operation condition adjustment scheme and a priority scheduling scheme based on the information; receiving a user configuration signal to determine a calibration operation type, and dynamically correcting the initial calibration strategy set according to the calibration operation type; the corrected strategy is executed, and the water pump dynamic characteristic model is updated based on execution result feedback. According to the method, collaborative implementation of model calibration and efficiency optimization in the operation process of the water pump is achieved.

Description

Technical Field

[0001] This invention relates to the field of water pump energy efficiency optimization technology, specifically a method for dynamic model calibration and efficiency optimization of water pumps. Background Technology

[0002] As a core component of fluid transport systems, water pumps are widely used in industrial circulation, urban water supply, agricultural irrigation, and many other fields. Their operating status directly affects the system's energy consumption, stability, and service life. In actual operation, water pumps often face complex and variable operating conditions: flow rate changes in real time due to fluctuations in end-user demand; pressure exhibits non-steady-state characteristics due to pipeline resistance; speed deviates due to performance degradation of the drive unit or load changes; and fluctuations in ambient temperature indirectly alter the physical properties of the medium, thus affecting the pump's output efficiency.

[0003] In traditional water pump operation and management, the models used are mostly static characteristic models. Their parameter settings are based on the rated operating conditions of the equipment at the time of manufacture or historical operating data, which makes it difficult to respond to the aforementioned dynamic changes in real time. For example, when the pipeline flow suddenly increases, the static model still calculates and outputs based on the preset parameters, which can easily lead to a significant deviation between the actual pressure and the model's predicted value. This not only causes energy waste but may also lead to safety hazards such as pipeline overpressure.

[0004] At the parameter analysis level, existing technologies often focus on the independent evaluation of single parameters, neglecting the inherent relationships between operational parameter nodes. For example, there is a nonlinear coupling relationship between flow rate and pressure; changes in rotational speed will simultaneously affect the dynamic response of both flow rate and pressure, while changes in ambient temperature may indirectly affect the operating resistance of the water pump by altering the viscosity of the medium. This neglect of the interrelationships between parameters makes the identification of key influencing parameters inaccurate, inefficient parameter combinations difficult to detect in a timely manner, and deviation nodes in the model unable to be accurately located, thus leading to a lack of targeted calibration operations.

[0005] The generation and execution of calibration strategies also have significant limitations. Existing methods often generate calibration strategies that are merely single parameter compensation schemes or simple adjustments for specific operating conditions, lacking a comprehensive consideration of parameter compensation, operating condition adjustment, and priority scheduling. Furthermore, the generation of calibration strategies rarely incorporates the user's actual configuration requirements. The different priorities users have regarding pump operation in various scenarios (e.g., some scenarios prioritize energy consumption, while others prioritize stability) cannot be effectively integrated, resulting in limited strategy applicability. In addition, most calibration processes lack post-execution feedback mechanisms, preventing timely feedback of calibration results to model updates. This causes the model to consistently lag behind actual operating conditions, creating an inefficient cycle of "calibration-deviation-recalibration." Summary of the Invention

[0006] The purpose of this invention is to provide a method for calibrating and optimizing the dynamic model of a water pump to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides a method for dynamic model calibration and efficiency optimization of a water pump, the method comprising: Step 1: Construct a dynamic characteristic model of the water pump and update the dynamic characteristic model of the water pump based on real-time operating data; wherein, the dynamic characteristic model of the water pump includes multiple operating parameter nodes, the correlation between parameter nodes and the characteristic information of each parameter node, and the real-time operating data includes flow rate change, pressure fluctuation, speed feedback and ambient temperature data. Step 2: Identify key influencing parameters and inefficient parameter combinations in the dynamic characteristic model of the water pump using parameter sensitivity analysis algorithms, and confirm deviation nodes. Step 3: Generate an initial calibration strategy set based on the key influencing parameters and deviation nodes; the initial calibration strategy set includes parameter compensation schemes, operating condition adjustment schemes, and priority scheduling schemes. Step 4: Receive the user configuration signal, determine the calibration operation type based on the user configuration signal, and dynamically modify the initial calibration strategy set based on the calibration operation type; Step 5: Execute the dynamically corrected initial calibration strategy set, and update the dynamic characteristic model of the water pump based on the execution result feedback.

[0008] Preferably, the specific implementation of constructing the dynamic characteristic model of the water pump in step 1 includes: Step 1.1: Collect historical operating data, including pump model parameters, pipeline characteristic curves, historical fault records, and energy efficiency test reports; Step 1.2: Extract multi-dimensional features from the historical operating data to generate feature information for parameter nodes; the feature information includes parameter type, adjustment sensitivity, response delay coefficient, operating stability index, historical calibration deviation rate, and dynamic correction coefficient; wherein: The response delay coefficient is dynamically assigned based on the signal transmission distance, sensor sampling frequency, and control algorithm complexity to obtain the corresponding response delay coefficient; Preferably, the dynamic correction coefficient is composed of a weighted sum of three parts, specifically including: Divide the historical calibration deviation rate of the parameter node by the preset maximum deviation rate benchmark value, and then multiply by the correction adjustment factor to obtain the historical calibration deviation rate percentage, which is used as the first part. Divide the current response delay coefficient by the average delay coefficient, and then multiply by the correction adjustment factor to obtain the current response delay percentage, which is the second part. Divide the operational stability index by the maximum stability threshold, and then multiply it by the correction adjustment factor to obtain the operational stability percentage, which is the third part. The dynamic correction coefficient is obtained by adding the first part, the second part, and the third part together. Step 1.3: Construct the topology of the dynamic characteristic model of the water pump based on data modeling tools; in the topology: Parameter nodes are connected by directed edges, which represent the direction of parameter influence and the strength of the association. Each parameter node is bound to the feature information, forming a complete node description that includes parameter type label, adjustment sensitivity value, response delay coefficient, operational stability index, historical calibration deviation rate, and dynamic correction coefficient.

[0009] Preferably, the association strength in step 1.3 is specifically as follows: Step 1.3.1: Based on the historical operation data collected in Step 1.1, calculate the frequency and magnitude of parameter interactions between adjacent parameter nodes; Step 1.3.2: Calculate the association strength value based on the parameter interaction frequency, parameter interaction amplitude, and node feature information generated in Step 1.2; the calculation process of the association strength value is as follows; Divide the parameter interaction frequency by the set maximum interaction frequency baseline value to obtain the normalized interaction frequency ratio. Divide the parameter interaction amplitude by the set maximum interaction amplitude baseline value to obtain the normalized interaction amplitude ratio. The average value of the regulation sensitivity of the source node and the target node is taken as the contribution value of the regulation sensitivity to the correlation strength. The normalized interaction frequency ratio, the normalized interaction amplitude ratio, and the contribution of the adjusted sensitivity to the association strength are multiplied by preset weight coefficients, and the association strength value is the sum of the three parts, provided that the sum of the weight coefficients is 1. Step 1.3.3: Assign weights to directed edges based on their association strength values: If the association strength value is greater than or equal to the first threshold, then a high association strength label is assigned to the directed edge; If the association strength value is between the second threshold and the first threshold, then it is the association strength label in the directed edge assignment; If the association strength value is less than the second threshold, then a low association strength label is assigned to the directed edge.

[0010] Preferably, step 2 specifically includes the following sub-steps: Step 2.1: Use the Monte Carlo simulation algorithm to traverse the dynamic characteristic model of the water pump and extract all complete parameter transfer paths from input parameters to output energy efficiency; Step 2.2: Calculate the energy conversion efficiency of each parameter transmission path; Step 2.3: Label key influencing parameters and inefficient parameter combinations based on energy efficiency conversion efficiency; specifically including the following sub-steps: The energy conversion efficiency value is compared with a preset efficiency threshold range, which includes a high efficiency threshold and a low efficiency threshold. If the energy efficiency conversion efficiency value is greater than or equal to the high efficiency threshold, the parameter in the corresponding parameter transmission path will be marked as a key influencing parameter. If the energy conversion efficiency value is less than the inefficient threshold, the parameter combination in the corresponding parameter transmission path will be marked as an inefficient parameter combination. Step 2.4: Perform differential analysis on key influencing parameters and combinations of inefficient parameters, and identify deviation nodes; this includes the following sub-steps: Extract the association edges with the highest association strength values ​​from the key influencing parameters, mark them as core parameter links, and assign additional parameter monitoring frequencies to the core parameter links; For each inefficient parameter combination, the dynamic correction coefficients of all nodes in the combination are extracted, and the node with the lowest dynamic correction coefficient is selected. If the dynamic correction coefficients of multiple nodes are the same and all are the lowest values, their historical calibration deviation rates are further compared, and the node with the highest deviation rate is identified as the deviation node. The core parameter link information of key influencing parameters and the deviation node information of inefficient parameter combinations are associated and stored.

[0011] Preferably, the calculation of energy conversion efficiency in step 2.2 includes the following sub-steps: Extract the dynamic correction coefficients of all nodes in the parameter transmission path and calculate the geometric mean of the node dynamic correction coefficients; Extract the weight labels of all associated edges in the parameter transmission path and calculate the arithmetic mean of the association strength values; The statistical path transmission delay is defined as the sum of the response delays between adjacent nodes in the path. Energy conversion efficiency is calculated using the following comprehensive evaluation logic: The comprehensive evaluation value is obtained by multiplying the geometric mean of the node dynamic correction coefficients with the arithmetic mean of the association strength values; The path propagation delay is added to a preset zero-prevention constant, which is a very small positive number used to avoid calculation errors. Divide the comprehensive evaluation value by the delay adjustment value to obtain the energy efficiency conversion efficiency value; the higher the energy efficiency conversion efficiency value, the better the energy efficiency conversion capability and stability of the parameter transmission path.

[0012] Preferably, the specific implementation of generating the initial calibration strategy set in step 3 includes: Step 3.1: Generate parameter compensation schemes for inefficient parameter combinations; specifically including the following sub-steps: Extract the inefficient parameter combinations marked in steps 2.3 and 2.4 and their corresponding deviation node information; Traverse the adjacent nodes of the deviation node and filter out the adjacent nodes whose compensation margin meets the preset conditions; the specific filtering process is that the current parameter deviation rate of the adjacent node is lower than the preset deviation threshold and its adjustment sensitivity is greater than the current disturbance value. Calculate the parameter compensation amount of the deviation node, that is, the difference between the current measured value and the model predicted value, based on the compensation margin and correlation strength value of the adjacent nodes; Step 3.2: Generate an operational condition adjustment plan for the core parameter link; specifically including the following sub-steps: Extract the key influencing parameters and their core parameter link information marked in step 2.3; Obtain the current operating parameters and ambient temperature data of the core parameter link; If the current operating parameters exceed the preset operating threshold or the ambient temperature risk level is higher than the preset risk threshold, the following adjustments will be made: The implementation of the operating condition optimization strategy is as follows: Using the topology described in 1.3, search for parameter combinations in alternative operating conditions that have a correlation strength value no lower than the original operating condition and a higher energy conversion efficiency; if no alternative operating condition can be found, retain the original operating condition, but trigger an early warning signal and generate an operating condition optimization suggestion report, which includes increasing the cooling system power, temporarily activating backup parameter nodes, or adjusting the operating load; Implement risk aversion strategies, specifically: Based on the pump model parameters and the historical number of anomalies and total running time recorded in the pipeline characteristic curves collected in step 1.1, the risk coefficient of the nodes in the working condition is calculated. Select nodes with risk coefficients lower than a preset threshold and compatible parameter characteristics as backup nodes; Replace high-risk nodes with backup nodes, and recalculate the correlation strength value and energy efficiency conversion efficiency of the operating conditions after replacement; Optimized operating condition commands are generated and added to the initial calibration strategy set.

[0013] Preferably, in step 1.2, the response delay coefficient is dynamically assigned a value based on the signal transmission distance, sensor sampling frequency, and control algorithm complexity to obtain the corresponding response delay coefficient. The specific process is as follows: The signal transmission distance is divided into three levels: long, medium, and short, and the distance weight is quantified based on the type of transmission medium. The sensor sampling frequency is divided into three categories: high frequency, medium frequency, and low frequency, and the frequency weight is assigned according to the data accuracy requirements, signal noise level, and hardware performance. The response latency coefficient is calculated using the following latency coefficient generation rules: Assign basic weights to signal transmission distance and sensor sampling frequency respectively; The response delay coefficient is generated by weighting and summing the basic weights and the complexity coefficients of the control algorithm.

[0014] Preferably, the calibration operation types in step 4 include parameter node update, correlation adjustment, and dynamic optimization of correction coefficients; Step 4, which generates the target calibration strategy set, specifically includes the following steps: Step 4.1: If the calibration operation type is parameter node update, recalculate the dynamic correction coefficient of the affected node and update the parameter compensation scheme. Step 4.2: If the calibration operation type is correlation adjustment, adjust the correlation strength value and re-evaluate the core parameter link to generate a new operating condition adjustment plan; Step 4.3: If the calibration operation type is dynamic optimization of correction coefficients, the original dynamic correction coefficients are overwritten based on the direct correction value input by the user, and the priority scheduling scheme is corrected synchronously.

[0015] Preferably, step 5 specifically includes the following sub-steps: Step 5.1: Execute the dynamically corrected initial calibration strategy set, including parameter compensation scheme, operating condition adjustment scheme and priority scheduling scheme; Step 5.2: Collect feedback data in real time during the execution process. The feedback data includes parameter compensation completion rate, actual response time of operating condition adjustment, energy efficiency index compliance rate, and node operation stability change value. Step 5.3: Adjust the dynamic correction coefficients of the parameter nodes in the dynamic characteristic model of the water pump; Step 5.4: Update the topology and correlation strength values ​​of the pump dynamic characteristic model based on the updated dynamic correction coefficients.

[0016] Compared with the prior art, the beneficial effects of the present invention are: This dynamic model calibration and efficiency optimization method for water pumps overcomes the limitations of traditional static models by constructing a dynamically updated characteristic model. The model is continuously updated based on real-time operating data, capturing the dynamic characteristics of parameters such as flow rate changes, pressure fluctuations, speed feedback, and ambient temperature in real time. This ensures the model remains consistent with the actual operating state of the water pump, avoiding calibration deviations caused by model lag. This dynamic characteristic model no longer relies on fixed parameters or historical data, but rather realistically reflects the interaction of parameters under different operating conditions through the correlation and characteristic information between parameter nodes, providing a practical basis for subsequent parameter analysis and calibration strategy generation.

[0017] The application of parameter sensitivity analysis algorithms solves the problem of inaccurate parameter identification in existing technologies. This algorithm can not only identify key influencing parameters but also discover inefficient parameter combinations and confirm deviation nodes, thanks to its in-depth analysis of the correlations between parameter nodes. By comprehensively considering the coupling effects of parameters such as flow rate, pressure, speed, and temperature, the algorithm can accurately determine which parameters have the greatest impact on pump operating efficiency, which parameter combinations lead to increased energy consumption, and which nodes' outputs deviate from actual values. This precise identification capability makes calibration operations no longer blind but can directly target core issues, fundamentally improving the effectiveness of calibration.

[0018] The initial calibration strategy set integrates parameter compensation, operating condition adjustment, and priority scheduling schemes, breaking through the limitations of traditional single strategies. Parameter compensation can directly correct the output of deviation nodes, operating condition adjustment can reduce parameter fluctuations by changing operating conditions, and priority scheduling can rationally allocate resources according to the needs of different scenarios. The combination of these three forms a multi-dimensional calibration system. This system can cope with different types of operating deviations: when there is a fixed deviation in parameters, the parameter compensation scheme can quickly correct it; when operating conditions change drastically, the operating condition adjustment scheme can adapt in a timely manner; when there are multiple calibration requirements, the priority scheduling scheme can ensure that critical requirements are met first.

[0019] Dynamically adjusting the calibration strategy based on user-configured signals enhances the method's adaptability to various scenarios. Different users have different requirements for pump operation in different scenarios. Some may be more concerned with short-term stability, others may focus more on long-term energy consumption control, and still others may need to prioritize ensuring that a certain parameter meets the standard within a specific time period. By receiving and responding to user-configured signals, the calibration strategy can adjust its direction and focus according to these specific needs. This transforms the strategy from a generalized template into a truly personalized solution tailored to the user's actual needs, thereby achieving optimal results in various application scenarios.

[0020] The feedback and update mechanism for execution results forms a closed loop of continuous optimization. After the calibration strategy is executed, its effects are captured in real time and used to update the dynamic characteristic model. This closed-loop design allows the model to continuously absorb calibration experience, gradually improving its ability to predict and simulate the pump's operating status. As operating time accumulates, the model becomes increasingly accurate, the results of parameter sensitivity analysis become more reliable, and the generation and correction of calibration strategies become more efficient. This continuous optimization process allows the pump's operating status to continuously evolve towards a better direction, reducing repeated calibrations caused by the model's disconnect from reality, lowering maintenance costs, and ensuring that the pump maintains a high level of efficiency throughout long-term operation. Attached Figure Description

[0021] Figure 1 This is a schematic diagram illustrating the working principle of the water pump dynamic model calibration and efficiency optimization method described in this invention. Figure 2 A flowchart for constructing a dynamic characteristic model of a water pump; Figure 3 The flowchart for parameter sensitivity analysis; Figure 4 This is a flowchart for generating the initial calibration strategy set. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] Please see Figure 1 This invention provides a method for calibrating and optimizing the dynamic model of a water pump, the method comprising: Step 1: Construct a dynamic characteristic model of the water pump and update the dynamic characteristic model of the water pump based on real-time operating data; wherein, the dynamic characteristic model of the water pump includes multiple operating parameter nodes, the correlation between parameter nodes and the characteristic information of each parameter node, and the real-time operating data includes flow rate change, pressure fluctuation, speed feedback and ambient temperature data.

[0024] Step 2: Identify key influencing parameters and inefficient parameter combinations in the dynamic characteristic model of the water pump using parameter sensitivity analysis algorithms, and confirm deviation nodes.

[0025] Step 3: Generate an initial calibration strategy set based on the key influencing parameters and deviation nodes; the initial calibration strategy set includes parameter compensation schemes, operating condition adjustment schemes, and priority scheduling schemes.

[0026] Step 4: Receive the user configuration signal, determine the calibration operation type based on the user configuration signal, and dynamically modify the initial calibration strategy set based on the calibration operation type.

[0027] Step 5: Execute the dynamically corrected initial calibration strategy set, and update the dynamic characteristic model of the water pump based on the execution result feedback.

[0028] Example 1: See Figure 2In constructing a dynamic characteristic model of the water pump and updating the model based on real-time operating data, the first step is to collect historical operating data. This historical operating data includes pump model parameters, pipeline characteristic curves, historical fault records, and energy efficiency test reports. Specifically, the pump model parameters include inherent attributes such as rated power, head, flow range, and speed; the pipeline characteristic curves reflect the relationship between flow rate and pressure loss in the pipeline system; the historical fault records detail the types of faults that occurred during past operation, their occurrence time, duration, and handling methods; and the energy efficiency test reports record the pump's energy efficiency level data under different operating conditions.

[0029] Multi-dimensional feature extraction is performed on the collected historical operating data to generate feature information for parameter nodes. This feature information includes parameter type, regulation sensitivity, response delay coefficient, operational stability index, historical calibration deviation rate, and dynamic correction coefficient. Parameter types are categorized based on their role in pump operation, such as flow parameters, pressure parameters, speed parameters, and temperature parameters. Regulation sensitivity characterizes the rate of change of a parameter under external regulation; its value is derived by analyzing the parameter's response to regulation signals in historical data. Operational stability index is determined by statistically analyzing the parameter's fluctuation amplitude over a period of time; the smaller the fluctuation amplitude, the higher the operational stability index. The historical calibration deviation rate is the statistical result of the deviation between the actual value and the calibration target value of the parameter during historical calibration.

[0030] The response delay coefficient is dynamically assigned based on the signal transmission distance, sensor sampling frequency, and control algorithm complexity. The signal transmission distance is divided into three levels: long, medium, and short, each corresponding to a different base value. This base value is also adjusted based on the transmission medium type; for example, the adjustment coefficient differs between fiber optic and cable transmission. The sensor sampling frequency is categorized into high, medium, and low frequencies, each with its own base value, which is then adjusted based on data accuracy requirements, signal noise levels, and hardware performance. The control algorithm complexity is determined by a coefficient based on factors such as the algorithm's computational steps and the number of logical branches. The response delay coefficient is obtained by weighting and summing the signal transmission distance and sensor sampling frequency coefficients with the control algorithm complexity coefficient.

[0031] The dynamic correction coefficient consists of a weighted sum of three parts. The first part is the historical calibration deviation rate percentage, calculated by dividing the historical calibration deviation rate of the parameter node by a preset maximum deviation rate benchmark value, and then multiplying by a correction adjustment factor. The preset maximum deviation rate benchmark value is determined based on the pump's operating requirements and the maximum deviation observed in historical data, while the correction adjustment factor is set according to the parameter's importance. The second part is the current response delay percentage, calculated by dividing the current response delay coefficient by the average delay coefficient, and then multiplying by the correction adjustment factor. The average delay coefficient is obtained by averaging the response delay coefficients of all parameter nodes. The third part is the operational stability percentage, calculated by dividing the operational stability index by the maximum stability threshold, and then multiplying by the correction adjustment factor. The maximum stability threshold is an upper limit set based on the pump's stable operation requirements. Summing these three parts yields the dynamic correction coefficient.

[0032] A topology for a dynamic characteristic model of a water pump is constructed using data modeling tools. In this topology, parameter nodes are interconnected by directed edges. The direction of the directed edges indicates the direction of influence between parameters; for example, changes in flow rate affect pressure, and the corresponding directed edge points from the flow rate node to the pressure node. The directed edges also reflect the strength of the correlation between parameter nodes. Each parameter node is bound to the extracted feature information, forming a complete node description. This means that each node not only contains basic features such as its parameter type and adjustment sensitivity, but also dynamic features such as response delay coefficients and dynamic correction coefficients, enabling the model to comprehensively reflect the dynamic characteristics of the water pump under different operating conditions. After the model is built, it is updated based on real-time operating data, including flow rate changes, pressure fluctuations, speed feedback, and ambient temperature data. By inputting this real-time data into the model, the feature information of the parameter nodes and the correlations between nodes are dynamically adjusted to ensure that the model accurately reflects the current operating state of the water pump.

[0033] Example 2: See Figure 3 When constructing the topology of a pump's dynamic characteristic model, determining the correlation strength between parameter nodes requires multiple steps. First, based on collected historical operating data, the frequency and magnitude of parameter interactions between adjacent parameter nodes are statistically analyzed. The parameter interaction frequency refers to the number of times parameter data is transmitted between two adjacent nodes per unit time, while the interaction magnitude is the amount of change in parameter data during transmission. For example, between a flow parameter node and a pressure parameter node, the number of times flow data is transmitted to the pressure node per hour is the interaction frequency, and the change in flow value during each transmission forms the basis for calculating the interaction magnitude.

[0034] Based on the above statistical results and node characteristic information, the correlation strength value is calculated. During the calculation, the parameter interaction frequency is first divided by a set maximum interaction frequency benchmark value to obtain a normalized interaction frequency ratio. The maximum interaction frequency benchmark value is determined based on the highest number of interactions that may occur under the maximum load condition of the water pump. Similarly, the parameter interaction amplitude is divided by a set maximum interaction amplitude benchmark value to obtain a normalized interaction amplitude ratio. The maximum interaction amplitude benchmark value is set with reference to the maximum parameter change observed in historical data. The average adjustment sensitivity of the source node and the target node is taken as the contribution value of the adjustment sensitivity to the correlation strength. These three results are multiplied by preset weighting coefficients, and the sum of the weighting coefficients is 1. The sum of the three is the correlation strength value.

[0035] Weight labels are assigned to directed edges based on their association strength values. If the association strength value is greater than or equal to a first threshold, a high association strength label is assigned to the directed edge; if the association strength value is between a second threshold and a first threshold, a medium association strength label is assigned; if the association strength value is equal to the second threshold, the same treatment applies as if the association strength value is between the second and first thresholds, assigning a medium association strength label; if the association strength value is less than the second threshold, a low association strength label is assigned. The first and second thresholds are determined by analyzing the impact of parameter associations on pump operating efficiency in historical data. A high association strength label indicates a significant parameter influence between the two nodes, while a low association strength label indicates a weak influence.

[0036] A Monte Carlo simulation algorithm is used to traverse the dynamic characteristic model of the water pump and extract all complete parameter transfer paths from input parameters to output energy efficiency. This algorithm simulates the transfer process under different parameter combinations through random sampling, ensuring coverage of all possible parameter transfer paths in the model. For example, starting with the input speed parameter, passing through intermediate parameter nodes such as flow rate and pressure, and finally reaching the output energy efficiency parameter, a complete transfer path is formed.

[0037] After calculating the energy conversion efficiency of each parameter transmission path, it is compared with a preset efficiency threshold range. The preset efficiency threshold range includes a high-efficiency threshold and a low-efficiency threshold, which are determined based on the energy efficiency performance of the water pump under standard operating conditions. If the energy conversion efficiency value is greater than or equal to the high-efficiency threshold, the parameter in the corresponding parameter transmission path is marked as a key influencing parameter; if the energy conversion efficiency value is less than the low-efficiency threshold, the parameter combination in the corresponding parameter transmission path is marked as an inefficient parameter combination.

[0038] Differential analysis is performed on combinations of key influencing parameters and inefficient parameters. For key influencing parameters, the correlation edges with the highest correlation strength are extracted and marked as core parameter links, and additional parameter monitoring frequencies are assigned to these core parameter links. The additional monitoring frequency is set based on the degree of impact of the core parameter links on the overall operation of the pump, and the monitoring frequency is higher than that of ordinary parameter nodes to monitor their operating status in real time.

[0039] For each inefficient parameter combination, the dynamic correction coefficients of all nodes in the combination are extracted, and the node with the lowest dynamic correction coefficient is selected. If multiple nodes have the same dynamic correction coefficient and are all at the lowest value, their historical calibration deviation rates are further compared, and the node with the highest deviation rate is identified as the deviation node. The core parameter link information of key influencing parameters and the deviation node information of inefficient parameter combinations are associated and stored. The stored information includes the node composition, association strength value, and monitoring frequency of the core parameter link, as well as the parameter type, dynamic correction coefficient, and historical calibration deviation rate of the deviation node, so as to be called when generating calibration strategies later.

[0040] Example 3: See Figure 4 When calculating the energy conversion efficiency of the parameter transmission path, the dynamic correction coefficients of all nodes in the path are first extracted, and the geometric mean of these coefficients is calculated. The geometric mean is calculated by multiplying all dynamic correction coefficients and then raising the result to the power of the number of nodes. Next, the weight labels of all associated edges in the path are extracted, and the arithmetic mean is calculated using the association strength values ​​corresponding to the weight labels. The arithmetic mean is the sum of all association strength values ​​divided by the number of associated edges. The path transmission delay is then calculated, defined as the sum of the response delays between adjacent nodes in the path. The response delay between each adjacent node is determined based on the node's response delay coefficient and the weight labels of the associated edges.

[0041] The energy conversion efficiency is calculated through a comprehensive evaluation using the following formula:

[0042] in, Indicates the energy conversion efficiency value; Represents the geometric mean of the node dynamic correction coefficients; This represents the arithmetic mean of the correlation strength values; Indicates path propagation delay; This represents the preset zero-prevention constant, which is a very small positive number.

[0043] When generating the initial calibration strategy set, to generate parameter compensation schemes for inefficient parameter combinations, it is necessary to first extract the marked inefficient parameter combinations and their corresponding deviation node information. The adjacent nodes of the deviation nodes are traversed, and adjacent nodes whose compensation margins meet preset conditions are selected. During the selection process, the current parameter deviation rate of the adjacent nodes must be lower than a preset deviation threshold, which is set according to the parameter type and operational requirements. Simultaneously, the adjustment sensitivity of the adjacent nodes must be greater than the current disturbance value, which is determined by the parameter fluctuation amplitude of the deviation node. The parameter compensation amount of the deviation node is calculated, i.e., the difference between the current measured value and the model predicted value. During the calculation, the compensation margin and correlation strength value of the adjacent nodes are referenced. The compensation margin is the range of parameters that the adjacent node can adjust; the higher the correlation strength value, the greater the compensation influence of the adjacent node on the deviation node.

[0044] When generating an operational condition adjustment plan for the core parameter link, the key influencing parameters and their core parameter link information are first extracted. The current operational condition parameters and ambient temperature data of the core parameter link are then obtained. The current operational condition parameters include real-time monitoring data such as flow rate, pressure, and speed, while the ambient temperature data is collected through temperature sensors installed around the water pump.

[0045] If the current operating parameters exceed the preset operating threshold or the ambient temperature risk level is higher than the preset risk threshold, an operating condition optimization strategy and a risk avoidance strategy will be executed. The operating condition optimization strategy searches for alternative operating conditions through the topology. The alternative operating conditions must meet the following requirements: the correlation strength value must be no lower than the original operating condition, and the energy conversion efficiency must be higher. If no alternative operating condition is found, the original operating condition is retained, and an early warning signal is triggered, and an operating condition optimization suggestion report is generated. The report includes the specific numerical range for increasing the cooling system power, the identifier of the temporarily activated backup parameter node, and the range of adjustment of the operating load.

[0046] The risk avoidance strategy calculates the risk coefficient of nodes in the operating condition based on the historical anomaly count and total runtime recorded in the collected pump model parameters and pipeline characteristic curves. The ratio of historical anomaly count to total runtime serves as the basis for risk coefficient calculation. Nodes with risk coefficients below a preset threshold and compatible parameter characteristics are selected as backup nodes. Parameter compatibility means that the parameter type, adjustment range, etc., of the backup node match those of the original node. After replacing high-risk nodes with backup nodes, the correlation strength value and energy conversion efficiency of the operating condition after replacement are recalculated, using the same calculation method as before. An optimized operating condition instruction is generated, including the target values ​​and adjustment steps for each parameter, and this instruction is added to the initial calibration strategy set.

[0047] Example 4: The determination of the response delay coefficient requires dynamic assignment based on the signal transmission distance, sensor sampling frequency, and control algorithm complexity. The signal transmission distance is divided into three levels: far, medium, and near, each corresponding to a different base value. The preset base weights for signal transmission distance are: far level = 0.6, medium level = 0.3, and near level = 0.1. The transmission medium type correction coefficients are: fiber optic transmission = 1.0 (no attenuation), cable transmission = 0.8 (slight attenuation), and wireless transmission = 0.6 (significant attenuation). The final signal transmission distance weight = base weight × transmission medium type correction coefficient. For example, when a signal is transmitted between a water pump and a remote monitoring center, a distance exceeding 500 meters is classified as far; a distance between 100 and 500 meters is classified as medium; and a distance less than 100 meters is classified as near. Simultaneously, the distance weight is quantified based on the transmission medium type; if the transmission medium is fiber optic, its weight value is relatively high; if it is ordinary cable, its weight value is relatively low.

[0048] Sensor sampling frequencies are categorized into three types: high frequency, medium frequency, and low frequency. A sensor sampling more than 1000 times per second is considered high frequency; between 100 and 1000 times per second is considered medium frequency; and less than 100 times per second is considered low frequency. Frequency weights are assigned based on data accuracy requirements, signal noise levels, and hardware performance. For scenarios requiring high-precision data, low signal noise, and strong hardware performance, high-frequency sampling frequencies have a higher weight; conversely, the weight decreases accordingly.

[0049] Basic weights are assigned to signal transmission distance and sensor sampling frequency, respectively, based on their impact on response delay. The preset basic weights for sensor sampling frequency are: high frequency = 0.5, mid-frequency = 0.3, and low frequency = 0.2. For data accuracy requirements, the correction coefficients are: high accuracy (error ≤ 0.5%) = 1.2, mid-accuracy (error 0.5%-2%) = 1.0, and low accuracy (error > 2%) = 0.8. For signal noise level correction coefficients, the correction coefficient is: low noise (signal-to-noise ratio ≥ 40dB) = 1.1, mid-noise (signal-to-noise ratio ≥ 20dB) = 1.1, and mid-noise (signal-to-noise ratio ≥ 20dB) = 1.1. The correction coefficients are as follows: B-40dB) correction coefficient = 1.0, high noise (signal-to-noise ratio <20dB) correction coefficient = 0.9; hardware performance correction coefficients: high-performance hardware (response time ≤10ms) correction coefficient = 1.0, ordinary hardware (response time 10ms-50ms) correction coefficient = 0.9, low-performance hardware (response time >50ms) correction coefficient = 0.8; final sensor sampling frequency weight = basic weight × data accuracy requirement correction coefficient × signal noise level correction coefficient × hardware performance correction coefficient. The basic weight and control algorithm complexity coefficient are weighted and summed to generate the response delay coefficient. The preset signal transmission distance weight is... ( =0.4), the weighting coefficient of the sensor sampling frequency weight is ( =0.3), the weighting coefficient for controlling the algorithm complexity coefficient is ( =0.3), and satisfies The algorithm complexity coefficient is set according to the number of algorithm steps: 0.2 for simple algorithms (≤10 steps), and 0.2 for medium algorithms (11 steps). For algorithms with ≤30 steps, the coefficient is 0.5; for complex algorithms (>30 steps), the coefficient is 0.8. The final response delay coefficient is calculated as follows: (signal transmission distance weight × ...) ) + (sensor sampling frequency weight × ) + (Control algorithm complexity coefficient × The complexity coefficient of an algorithm is determined by the number of computational steps and the level of logical judgments it contains. The more steps and the more complex the levels, the larger the coefficient value.

[0050] Calibration operation types include parameter node updates, correlation adjustments, and dynamic optimization of correction coefficients. When the calibration operation type is a parameter node update, the dynamic correction coefficients of the affected nodes are recalculated. For example, if a flow parameter node is updated due to sensor replacement, the dynamic correction coefficients of that node and its directly associated pressure and speed parameter nodes need to be recalculated. The calculation of dynamic correction coefficients involves re-evaluating and weighting the historical calibration deviation rate, current response delay coefficient, and operational stability indicators. After the calculation is completed, the parameter compensation scheme is updated, adjusting the compensation amounts and priorities related to these nodes in the original scheme to ensure that the compensation scheme matches the updated node characteristics.

[0051] When the calibration operation type is correlation adjustment, the correlation strength value is adjusted and the core parameter link is re-evaluated. For example, if pipeline modifications change the correlation between flow parameter nodes and pressure parameter nodes, the interaction frequency and amplitude of the parameters between the two need to be recalculated, and a new correlation strength value is calculated based on the adjustment sensitivity. Based on the new correlation strength value, the weight labels of the correlation edges are redefined. If the correlation strength value of a certain correlation edge in the original core parameter link decreases, it may lead to a change in the core parameter link, requiring relabeling of the new core parameter link. A new operating condition adjustment scheme is generated based on the new core parameter link, adjusting the operating condition parameter range and optimization strategy in the scheme to adapt to the change in correlation.

[0052] When the calibration operation type is dynamic optimization of correction coefficients, the direct correction value based on user input overrides the original dynamic correction coefficients. The direct correction value input by the user can be set according to field operating experience or special operating conditions. For example, in the initial stage of pump startup, the user inputs the direct correction value of a certain parameter node based on experience. After overriding the original dynamic correction coefficients, the priority scheduling scheme is adjusted synchronously. The priority scheduling scheme is used to determine the execution order of each calibration strategy. After the correction coefficients are adjusted, the calibration priority of relevant parameter nodes may change and needs to be reordered to ensure that the calibration strategy with higher priority is executed first.

[0053] Example 5: Execution of the initial calibration strategy set after dynamic correction, encompassing parameter compensation schemes, operating condition adjustment schemes, and priority scheduling schemes. The parameter compensation scheme involves adjusting parameters at deviation nodes. Based on the calculated compensation amount, the control system sends adjustment commands to the corresponding actuators, such as adjusting valve openings to change flow parameters or adjusting motor voltage to change speed parameters. The operating condition adjustment scheme, based on optimized operating condition commands, adjusts the overall operating parameters of the water pump, including setting new flow target values, pressure ranges, and speed ranges, while coordinating the operating status of related auxiliary equipment, such as starting the cooling system or switching to a standby pump unit. The priority scheduling scheme determines the execution order of each calibration operation, executing them sequentially from highest to lowest priority to ensure that the calibration of critical parameters is performed first and to avoid operational conflicts.

[0054] Real-time data collection during execution includes parameter compensation completion rate, actual response time for operating condition adjustments, energy efficiency index compliance rate, and node operational stability changes. The parameter compensation completion rate is the ratio of actual compensation to planned compensation, calculated by comparing measured parameter values ​​before and after adjustment. The actual response time for operating condition adjustments is the time interval from sending the adjustment command to the parameter reaching the target range, obtained by recording the time difference between the command sending moment and the parameter stabilization moment. The energy efficiency index compliance rate is the ratio of the actual energy efficiency index reaching the preset standard to the total operating time; energy efficiency indicators are measured in real-time using dedicated energy efficiency monitoring instruments. Node operational stability changes are calculated by comparing the fluctuation amplitude of parameters before and after adjustment, with the fluctuation amplitude processed using the standard deviation statistical method.

[0055] Adjust the dynamic correction coefficients of the parameter nodes in the pump's dynamic characteristic model. Recalculate each component of the dynamic correction coefficients based on feedback data such as the parameter compensation completion rate and node operational stability changes. If the parameter compensation completion rate is low, it indicates a significant impact from the historical calibration deviation rate, requiring a recalculation of the historical calibration deviation rate percentage. If the node operational stability changes show increased fluctuations, the operational stability percentage needs corresponding adjustment. During recalculation, maintain the weighting method for each component, updating only the input base data. For example, the historical calibration deviation rate uses the latest compensation deviation data, and the operational stability index uses the adjusted fluctuation statistics.

[0056] The topology and correlation strength values ​​of the pump's dynamic characteristic model are updated based on the updated dynamic correction coefficients. The topology update includes refreshing the parameter node feature information and writing the adjusted dynamic correction coefficients, response delay coefficients, etc., into the description information of the corresponding nodes. The correlation strength value update requires recalculating the interaction frequency and amplitude between parameter nodes and calculating new correlation strength values ​​based on the updated adjustment sensitivity. The weight labels of directed edges are reassigned according to the new correlation strength values; if the correlation strength value of a certain edge increases, its label may be adjusted from low correlation strength to medium correlation strength, and vice versa. The updated topology more accurately reflects the pump's current operating characteristics and the correlation relationships between parameters, providing a foundation for the next model calibration and efficiency optimization.

[0057] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0058] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for calibrating and optimizing the efficiency of a water pump dynamic model, characterized in that, Includes the following steps: Step 1: Construct a dynamic characteristic model of the water pump and update the dynamic characteristic model of the water pump based on real-time operating data; wherein, the dynamic characteristic model of the water pump includes multiple operating parameter nodes, the correlation between parameter nodes and the characteristic information of each parameter node, and the real-time operating data includes flow rate change, pressure fluctuation, speed feedback and ambient temperature data. Step 2: Identify key influencing parameters and inefficient parameter combinations in the dynamic characteristic model of the water pump using parameter sensitivity analysis algorithms, and confirm deviation nodes. Step 3: Generate an initial calibration strategy set based on the key influencing parameters and deviation nodes; the initial calibration strategy set includes parameter compensation schemes, operating condition adjustment schemes, and priority scheduling schemes. Step 4: Receive the user configuration signal, determine the calibration operation type based on the user configuration signal, and dynamically modify the initial calibration strategy set based on the calibration operation type; Step 5: Execute the dynamically corrected initial calibration strategy set, and update the dynamic characteristic model of the water pump based on the execution result feedback.

2. The method for calibrating and optimizing the dynamic model of a water pump according to claim 1, characterized in that: The specific implementation of constructing the dynamic characteristic model of the water pump in step 1 includes: Step 1.1: Collect historical operating data, including pump model parameters, pipeline characteristic curves, historical fault records, and energy efficiency test reports; Step 1.2: Extract multi-dimensional features from the historical operating data to generate feature information for parameter nodes; the feature information includes parameter type, adjustment sensitivity, response delay coefficient, operating stability index, historical calibration deviation rate, and dynamic correction coefficient; wherein: The response delay coefficient is dynamically assigned based on the signal transmission distance, sensor sampling frequency, and control algorithm complexity to obtain the corresponding response delay coefficient.

3. The method for calibrating and optimizing the dynamic model of a water pump according to claim 2, characterized in that, The dynamic correction coefficient is composed of a weighted sum of three parts, specifically including: Divide the historical calibration deviation rate of the parameter node by the preset maximum deviation rate benchmark value, and then multiply by the correction adjustment factor to obtain the historical calibration deviation rate percentage, which is used as the first part. Divide the current response delay coefficient by the average delay coefficient, and then multiply by the correction adjustment factor to obtain the current response delay percentage, which is the second part. Divide the operational stability index by the maximum stability threshold, and then multiply it by the correction adjustment factor to obtain the operational stability percentage, which is the third part. The dynamic correction coefficient is obtained by adding the first part, the second part, and the third part together. Step 1.3: Construct the topology of the dynamic characteristic model of the water pump based on data modeling tools; in the topology: Parameter nodes are connected by directed edges, which represent the direction of parameter influence and the strength of the association. Each parameter node is bound to the feature information, forming a complete node description that includes parameter type label, adjustment sensitivity value, response delay coefficient, operational stability index, historical calibration deviation rate, and dynamic correction coefficient.

4. The method for calibrating and optimizing the dynamic model of a water pump according to claim 3, characterized in that: The correlation strength in step 1.3 is specifically as follows: Step 1.3.1: Based on the historical operation data collected in Step 1.1, calculate the frequency and magnitude of parameter interactions between adjacent parameter nodes; Step 1.3.2: Calculate the association strength value based on the parameter interaction frequency, parameter interaction amplitude, and node feature information generated in Step 1.2; the calculation process of the association strength value is as follows; Divide the parameter interaction frequency by the set maximum interaction frequency baseline value to obtain the normalized interaction frequency ratio. Divide the parameter interaction amplitude by the set maximum interaction amplitude baseline value to obtain the normalized interaction amplitude ratio. The average value of the regulation sensitivity of the source node and the target node is taken as the contribution value of the regulation sensitivity to the correlation strength. The normalized interaction frequency ratio, the normalized interaction amplitude ratio, and the contribution of the adjusted sensitivity to the association strength are multiplied by preset weight coefficients, and the association strength value is the sum of the three parts, provided that the sum of the weight coefficients is 1. Step 1.3.3: Assign weights to directed edges based on their association strength values: If the association strength value is greater than or equal to the first threshold, then a high association strength label is assigned to the directed edge; If the association strength value is between the second threshold and the first threshold, then it is the association strength label in the directed edge assignment; If the association strength value is less than the second threshold, then a low association strength label is assigned to the directed edge.

5. The method for calibrating and optimizing the dynamic model of a water pump according to claim 2, characterized in that: Step 2 specifically includes the following sub-steps: Step 2.1: Use the Monte Carlo simulation algorithm to traverse the dynamic characteristic model of the water pump and extract all complete parameter transfer paths from input parameters to output energy efficiency; Step 2.2: Calculate the energy conversion efficiency of each parameter transmission path; Step 2.3: Label key influencing parameters and inefficient parameter combinations based on energy efficiency conversion efficiency; specifically including the following sub-steps: The energy conversion efficiency value is compared with a preset efficiency threshold range, which includes a high efficiency threshold and a low efficiency threshold. If the energy efficiency conversion efficiency value is greater than or equal to the high efficiency threshold, the parameter in the corresponding parameter transmission path will be marked as a key influencing parameter. If the energy conversion efficiency value is less than the inefficient threshold, the parameter combination in the corresponding parameter transmission path will be marked as an inefficient parameter combination. Step 2.4: Perform differential analysis on key influencing parameters and combinations of inefficient parameters, and identify deviation nodes; this includes the following sub-steps: Extract the association edges with the highest association strength values ​​from the key influencing parameters, mark them as core parameter links, and assign additional parameter monitoring frequencies to the core parameter links; For each inefficient parameter combination, the dynamic correction coefficients of all nodes in the combination are extracted, and the node with the lowest dynamic correction coefficient is selected. If the dynamic correction coefficients of multiple nodes are the same and all are the lowest values, their historical calibration deviation rates are further compared, and the node with the highest deviation rate is identified as the deviation node. The core parameter link information of key influencing parameters and the deviation node information of inefficient parameter combinations are associated and stored.

6. The method for calibrating and optimizing the dynamic model of a water pump according to claim 5, characterized in that: The calculation of energy conversion efficiency in step 2.2 includes the following sub-steps: Extract the dynamic correction coefficients of all nodes in the parameter transmission path and calculate the geometric mean of the node dynamic correction coefficients; Extract the weight labels of all associated edges in the parameter transmission path and calculate the arithmetic mean of the association strength values; The statistical path transmission delay is defined as the sum of the response delays between adjacent nodes in the path. Energy conversion efficiency is calculated using the following comprehensive evaluation logic: The comprehensive evaluation value is obtained by multiplying the geometric mean of the node dynamic correction coefficients with the arithmetic mean of the association strength values; The path propagation delay is added to a preset zero-prevention constant, which is a very small positive number used to avoid calculation errors. Divide the comprehensive evaluation value by the delay adjustment value to obtain the energy efficiency conversion efficiency value; the higher the energy efficiency conversion efficiency value, the better the energy efficiency conversion capability and stability of the parameter transmission path.

7. The method for calibrating and optimizing the dynamic model of a water pump according to claim 5, characterized in that, The specific implementation of generating the initial calibration strategy set in step 3 includes: Step 3.1: Generate parameter compensation schemes for inefficient parameter combinations; specifically including the following sub-steps: Extract the inefficient parameter combinations marked in steps 2.3 and 2.4 and their corresponding deviation node information; Traverse the adjacent nodes of the deviation node and filter out the adjacent nodes whose compensation margin meets the preset conditions; the specific filtering process is that the current parameter deviation rate of the adjacent node is lower than the preset deviation threshold and its adjustment sensitivity is greater than the current disturbance value. Calculate the parameter compensation amount of the deviation node, that is, the difference between the current measured value and the model predicted value, based on the compensation margin and correlation strength value of the adjacent nodes; Step 3.2: Generate an operational condition adjustment plan for the core parameter link; specifically including the following sub-steps: Extract the key influencing parameters and their core parameter link information marked in step 2.3; Obtain the current operating parameters and ambient temperature data of the core parameter link; If the current operating parameters exceed the preset operating threshold or the ambient temperature risk level is higher than the preset risk threshold, the following adjustments will be made: The implementation of the operating condition optimization strategy is as follows: Using the topology structure from step 1.3, search for parameter combinations in alternative operating conditions that have a correlation strength value no lower than the original operating condition and a higher energy conversion efficiency; if no alternative operating condition can be found, retain the original operating condition, but trigger an early warning signal and generate an operating condition optimization suggestion report, which includes increasing the cooling system power, temporarily activating backup parameter nodes, or adjusting the operating load; Implement risk aversion strategies, specifically: Based on the pump model parameters and the historical number of anomalies and total running time recorded in the pipeline characteristic curves collected in step 1.1, the risk coefficient of the nodes in the working condition is calculated. Select nodes with risk coefficients lower than a preset threshold and compatible parameter characteristics as backup nodes; Replace high-risk nodes with backup nodes, and recalculate the correlation strength value and energy efficiency conversion efficiency of the operating conditions after replacement; Optimized operating condition commands are generated and added to the initial calibration strategy set.

8. The method for calibrating and optimizing the dynamic model of a water pump according to claim 7, characterized in that: In step 1.2, the response delay coefficient is dynamically assigned a value based on the signal transmission distance, sensor sampling frequency, and control algorithm complexity. The specific process for obtaining the corresponding response delay coefficient is as follows: The signal transmission distance is divided into three levels: long, medium, and short, and the distance weight is quantified based on the type of transmission medium. The sensor sampling frequency is divided into three categories: high frequency, medium frequency, and low frequency, and the frequency weight is assigned according to the data accuracy requirements, signal noise level, and hardware performance. The response latency coefficient is calculated using the following latency coefficient generation rules: Assign basic weights to signal transmission distance and sensor sampling frequency respectively; The response delay coefficient is generated by weighting and summing the basic weights and the complexity coefficients of the control algorithm.

9. The method for calibrating and optimizing the dynamic model of a water pump according to claim 2, characterized in that: The calibration operation types in step 4 include parameter node update, correlation adjustment, and dynamic optimization of correction coefficients. Step 4, which generates the target calibration strategy set, specifically includes the following steps: Step 4.1: If the calibration operation type is parameter node update, recalculate the dynamic correction coefficient of the affected node and update the parameter compensation scheme. Step 4.2: If the calibration operation type is correlation adjustment, adjust the correlation strength value and re-evaluate the core parameter link to generate a new operating condition adjustment plan; Step 4.3: If the calibration operation type is dynamic optimization of correction coefficients, the original dynamic correction coefficients are overwritten based on the direct correction value input by the user, and the priority scheduling scheme is corrected synchronously.

10. The method for calibrating and optimizing the dynamic model of a water pump according to claim 2, characterized in that: Step 5 specifically includes the following sub-steps: Step 5.1: Execute the dynamically corrected initial calibration strategy set, including parameter compensation scheme, operating condition adjustment scheme and priority scheduling scheme; Step 5.2: Collect feedback data in real time during the execution process. The feedback data includes parameter compensation completion rate, actual response time of operating condition adjustment, energy efficiency index compliance rate, and node operation stability change value. Step 5.3: Adjust the dynamic correction coefficients of the parameter nodes in the dynamic characteristic model of the water pump; Step 5.4: Update the topology and correlation strength values ​​of the pump dynamic characteristic model based on the updated dynamic correction coefficients.

Images