Active detection method for power supply carrying capacity of distribution network based on load dynamic model

Abstract

The application discloses a kind of based on load dynamic model's distribution network power supply carrying capacity active detection method, steps are as follows: setting distribution network backup power storage unit type and calling its characteristic quantity, based on the characteristic quantity, initial strategy configuration is completed detection;According to initial strategy configuration, power supply account information and voltage, current operating parameter are collected, and the programmable dynamic load model suitable for the storage unit type is constructed;Real-time operating parameter is input into model and carried out carrying capacity evaluation, and dynamic adjustment strategy is output;Optimize load array in combination with storage unit type, initial strategy and dynamic adjustment strategy, execute active loading detection, generate final evaluation report and visual display in man-machine interface.This method realizes the accurate, efficient detection of distribution network backup power carrying capacity, adapts to different storage units, can accurately simulate impact load, meets the requirements of regulations, and provides data support for distribution network operation and maintenance.

Description

Technical Field

[0001] This invention belongs to the field of power supply testing technology, specifically relating to an active detection method for the load-carrying capacity of power supply in distribution networks based on a load dynamic model. Background Technology

[0002] As a critical power supply guarantee during power system faults or maintenance, the backup power supply of the distribution network directly determines the stability and reliability of the grid's emergency power supply, making it a core component of the distribution network's safe operation system. With the continuous expansion of the power grid's construction scale and the increasingly diversified load demands of the distribution network, the power grid has placed higher requirements on the backup power supply's transient impact load tolerance and cyclic load carrying capacity, and has also imposed stringent technical standards on the accuracy and efficiency of backup power supply load capacity testing.

[0003] Currently, the load-carrying capacity testing of backup power supplies in distribution networks still relies on traditional methods, mainly divided into two categories: voltage internal resistance testing and battery capacity testing. Both methods have significant technical shortcomings and are no longer suitable for the testing needs of modern distribution networks. The voltage internal resistance testing method relies solely on manual experience and simple measuring tools to collect basic voltage and internal resistance data of the backup power supply to determine its operating status. However, backup power supplies are normally in a standby floating charging state, and voltage data alone cannot accurately reflect their actual load-carrying capacity. Internal resistance data can only preliminarily detect the health status of the energy storage unit and cannot accurately determine its ability to withstand sudden load surges under scenarios such as switching operations and sudden faults. This results in numerous blind spots and is prone to causing power supply failures during emergency power supply phases due to misjudgments. While the battery capacity testing method can detect the actual capacity of the energy storage unit, its testing process is cumbersome, inefficient, and time-consuming. Furthermore, the testing process requires shutting down the backup power supply, severely impacting the normal operation planning of the distribution network and failing to meet the large-scale, routine testing needs of the distribution network.

[0004] In summary, existing methods for detecting the load-carrying capacity of backup power supplies in distribution networks generally suffer from problems such as low detection accuracy, large detection blind spots, low detection efficiency, and inability to adapt to the detection requirements of transient impact loads. There is a lack of technical solutions that can achieve accurate and rapid detection of the load-carrying capacity of backup power supplies under all operating conditions without shutting down the system. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention proposes an active detection method for the load-carrying capacity of power supply in distribution networks based on a dynamic load model. This method is adaptable to different types of energy storage units and can simulate the actual load conditions of distribution networks. It achieves high precision, high efficiency, and high adaptability in the detection process, thus solving a pressing technical problem in the field of backup power supply detection in distribution networks.

[0006] In a first aspect, this invention proposes an active detection method for the load-carrying capacity of power distribution networks based on a dynamic load model, comprising:

[0007] Set the energy storage unit type of the power supply backup power supply, and retrieve the characteristic quantities of the corresponding energy storage unit type; complete the initial strategy configuration for the load capacity detection of the power supply backup power supply based on the characteristic quantities; the characteristic quantities include the charging and discharging characteristics, internal resistance variation law, impact load tolerance threshold, capacity decay characteristics and constant current output characteristics of each energy storage unit type.

[0008] Based on the initial strategy configuration, the ledger information and voltage and current operating parameters of the backup power supply of the distribution network are collected, and a programmable dynamic load model adapted to this type of energy storage unit is constructed.

[0009] The operating parameters such as voltage and current of the backup power supply in the distribution network are collected in real time and input into the programmable dynamic load model to evaluate the load capacity and output a dynamic adjustment strategy for the load capacity.

[0010] Based on the energy storage unit type, initial strategy configuration and dynamic adjustment strategy of load capacity, the load array is optimized. Active load detection is performed on the optimized backup power supply to generate a final evaluation report of the load capacity of the distribution network backup power supply, which is then visualized on the human-machine interface.

[0011] This invention presents an active detection method for the load-carrying capacity of power distribution networks. Using the characteristic quantities of different energy storage unit types as the core adaptation basis, it first completes the customized configuration of the initial detection strategy. Then, based on this configuration, it collects power supply records and operating parameters to construct a programmable dynamic load model adapted to the energy storage unit type. By inputting real-time operating parameters into the model, it completes an accurate assessment of the load-carrying capacity and outputs a dynamic adjustment strategy. Furthermore, it optimizes the load array by combining the energy storage unit type, initial strategy configuration, and dynamic adjustment strategy, performing active loading detection on the power distribution network backup power supply. Finally, it generates a load-carrying capacity assessment report and visualizes it on a human-machine interface. The entire method achieves a closed-loop process from detection strategy customization, model construction, capacity assessment to active loading detection. Relying on the differentiated adaptation of energy storage unit characteristic quantities and the accurate simulation of the programmable dynamic load model, it solves the problems of low accuracy, poor efficiency, and inability to accurately detect impact load capacity in traditional detection methods. It achieves high-precision, high-efficiency, and full-condition active detection of the load-carrying capacity of power distribution network backup power supplies. The detection results are intuitive and verifiable, providing accurate data support for the safe operation and maintenance of power distribution network backup power supplies.

[0012] Preferably, the initial strategy configuration for detecting the load-carrying capacity of the distribution network backup power supply based on the feature quantity includes:

[0013] Extract the core parameters of the characteristic quantities corresponding to the energy storage unit type to form a structured set of characteristic quantity parameters;

[0014] By integrating the aforementioned set of feature parameters into the core detection algorithm, targeted modifications are made to the dynamic load model construction algorithm, load capacity assessment algorithm, and load array control algorithm. The parameter identification logic, load assessment weights, and driving threshold boundaries of each algorithm are adjusted to ensure that the operational logic of the core detection algorithm is compatible with the inherent operating characteristics of the current energy storage unit. This approach avoids the accuracy deviation problem caused by general algorithms detecting different energy storage units from the algorithm's underlying layer, laying a foundation for an adaptable algorithm for subsequent model construction and capacity assessment.

[0015] Based on the sensitivity of energy storage unit parameters reflected by the aforementioned feature parameter set, differentiated voltage and current synchronous acquisition strategies are configured to determine A / D sampling accuracy, sampling rate, filtering method, signal transformation rules, and amplitude limiting protection threshold, adapting to the signal acquisition requirements of energy storage units. This ensures that the data acquisition process conforms to the signal characteristics of different energy storage units, effectively avoiding signal distortion caused by mismatch between acquisition parameters and energy storage units, and ensuring the accuracy and reliability of the data source for subsequent modeling and evaluation.

[0016] Based on the energy storage unit's impact load tolerance threshold and constant current output characteristics, the basic parameters for load loading are preset, and the initial loading amplitude, single loading duration, loading steps, loading interval, and switch opening and closing cycle loading simulation frequency are determined. The initial loading parameters do not exceed the safety tolerance boundary of the energy storage unit.

[0017] Based on the capacity decay characteristics and overcurrent and overvoltage tolerance thresholds of the energy storage unit, a multi-level safety protection strategy is set during the testing process, clarifying the load derating ratio, shutdown triggering conditions, and emergency handling rules under abnormal operating conditions; emergency handling can be carried out in a timely manner according to the status changes of the energy storage unit, ensuring the safe operation of the distribution network backup power supply and testing device throughout the entire testing process.

[0018] The optimized core detection algorithm, customized data acquisition strategy, pre-set load loading strategy, and established security protection strategy are integrated into a unified initial strategy set and embedded into the detection device's processing unit. This serves as the foundation for subsequent data acquisition, model building, load capacity assessment, and load array control. This standardization and systematization of the detection strategy ensures a unified execution benchmark for subsequent stages such as data acquisition, model building, and load array control. It avoids the chaos in the detection process caused by disconnected strategies in different stages, and improves the coherence and controllability of the detection process.

[0019] Preferably, the step of collecting the ledger information and voltage and current operating parameters of the distribution network backup power supply based on the initial strategy configuration, and constructing a programmable dynamic load model adapted to the energy storage unit type, includes:

[0020] A customized data acquisition strategy is used to collect basic ledger information and real-time operating parameters of the distribution network backup power supply, avoiding model distortion caused by missing data dimensions. The ledger information includes the rated voltage, rated current, rated capacity, operation and maintenance records, and historical load characteristics of the distribution network backup power supply. The real-time operating parameters are analog voltage and current signals synchronously collected by high-precision sensors.

[0021] The algorithm for constructing a dynamic load model optimized by the energy storage unit features in the initial strategy configuration is retrieved, the preset polynomial dynamic load basic model is loaded, a time series data cache array is created, the above-mentioned data sources are imported into the cache array, and data preprocessing is completed by sliding window mean filtering.

[0022] The load characteristic parameters of the backup power supply in the distribution network under the historical stable operation state are read. Combined with the pre-processed real-time data source, the recursive least squares (RLS) identification algorithm is used to iteratively solve the linear model in the preset polynomial dynamic load basic model. During the iteration process, the number of parameter identification iterations matching the energy storage unit type is set to obtain the optimal load characteristic parameters that are suitable for the current backup power supply of the distribution network under test.

[0023] Based on the optimal load characteristic parameters, dynamic load characteristics are calculated, and the model is dynamically corrected by incorporating energy storage unit characteristics to obtain the final programmable dynamic load model. The programmable dynamic load model can accurately reflect the core characteristics of the backup power supply of the tested distribution network, such as the load target value and dynamic constant current value, and can effectively simulate load changes under actual operating conditions of the distribution network, such as normal load, transient impact load, and cyclic load of switch opening and closing.

[0024] Furthermore, before importing the aforementioned data source into the cache array, the process also includes:

[0025] The acquired voltage and current analog signals are sequentially processed by signal transformation, low-pass filtering, and amplitude limiting protection. The backup power supply voltage is transformed into a standard linear voltage of 0~3.3V through linear resistor voltage division, linear opto-isolation, and operational amplification. Then, it is converted into a digital quantity by A / D conversion with the accuracy set in the initial strategy configuration, forming a structured detection data source.

[0026] Preferably, the step of performing dynamic load characteristic calculation based on the optimal load characteristic parameters includes:

[0027] Calculate the change value of dynamic load impedance and load change rate of the backup power supply of the distribution network over time. The dynamic load impedance reflects the impedance change law of the load under normal and transient conditions, and the load change rate reflects the dynamic change speed of the load.

[0028] Preferably, the load array optimization, which combines energy storage unit type, initial strategy configuration, and dynamic load capacity adjustment strategy, and the active load detection for the optimized backup power supply, includes:

[0029] Based on the characteristic boundary of the energy storage unit type, the load array control algorithm fixed in the initial strategy configuration, and the loading parameter requirements in the dynamic adjustment strategy of load capacity, the linear load array composed of series and parallel power transistors is dynamically combined and optimized to adjust the number of series and parallel power transistors and the connection method.

[0030] Based on the drive signal of the dynamic closed-loop control circuit, the optimized linear load array is driven to perform active loading detection on the power supply backup power supply without shutdown. During the detection process, the real-time response parameters such as voltage, current and temperature of the power supply backup power supply and the actual output load parameters of the linear load array are collected simultaneously to identify the load performance and response characteristics of the power supply backup power supply under different load conditions in real time.

[0031] Preferably, the active load detection of the optimized backup power supply further includes: comparing the collected response parameters with the multi-level safety protection thresholds set in the initial strategy configuration in real time during the load detection process; if the detected parameters exceed the warning threshold, immediately performing load derating operation according to the dynamic adjustment strategy of load capacity; if the parameters exceed the termination threshold, immediately triggering a load termination command, stopping the active load detection and generating abnormal warning information.

[0032] Secondly, based on the same inventive concept, this application also proposes an active detection circuit for the load-carrying capacity of power distribution networks based on a load dynamic model. The circuit adopts an active detection method for the load-carrying capacity of power distribution networks based on a load dynamic model as described in the first aspect.

[0033] Preferably, the active detection circuit for the load-carrying capacity of the power distribution network based on the load dynamic model includes:

[0034] Data acquisition circuit module: includes signal conversion circuit, low-pass filter circuit and amplitude limiting protection circuit; used for synchronous acquisition, conversion, filtering and protection of voltage and current signals, and outputting digital signals to the arithmetic processing circuit module;

[0035] The arithmetic processing circuit module includes peripheral basic circuits, A / D conversion circuits, chip main control circuits, and DAC interface driver circuits; the chip main control circuit has built-in data processing programs, dynamic load model algorithm programs, load capacity assessment algorithm programs, and logic control programs.

[0036] The load drive circuit module includes an amplifier circuit, a differential feedback circuit, and a comparator circuit. The amplifier circuit is composed of an operational amplifier, which amplifies the 0~3.3V low voltage signal input from the operational processing circuit module into a 0~12V drive signal. The differential feedback circuit acquires the real-time current signal of the linear load array circuit module and feeds it back to the inverting input of the comparator circuit. The comparator circuit compares the amplified drive signal with the current feedback signal in real time to form a dynamic closed-loop control circuit and outputs the drive signal to the linear load array circuit module.

[0037] The linear load array circuit module consists of several low-switching-loss high-speed IGBT power transistors, a manganese-copper shunt, and a current feedback circuit. Each IGBT power transistor is connected in parallel with a fast recovery full-current diode. Multiple IGBT power transistors are connected in series and parallel to form a programmable linear load array. The load current is linearly regulated by the gate signal. The manganese-copper shunt is connected in series with the load array and uses micro-voltage acquisition to achieve real-time load current detection. The current signal is then transmitted to the differential feedback circuit of the load drive circuit module via the current feedback circuit.

[0038] The human-computer interaction circuit module includes a TTL driver interface circuit and a touch serial port display circuit. The TTL driver interface circuit enables bidirectional data interaction with the arithmetic processing circuit module, receiving detection data, evaluation results, and control commands and transmitting them to the touch serial port display circuit. The touch serial port display circuit is a serial port display circuit with a touch screen, which completes the functions of setting detection parameters, real-time display of detection status, visualization of load capacity evaluation results, and issuing manual commands.

[0039] Preferably, the detection circuit further includes:

[0040] The AT32F chip in the computing circuit module also integrates a logic judgment program unit and a human-machine interface control program unit. The logic judgment program unit completes the threshold judgment of the running parameters and the logic triggering of abnormal working conditions during the detection process. The human-machine interface control program unit realizes the instruction parsing, status feedback and dynamic interface update of the touch serial port display circuit.

[0041] Thirdly, based on the same inventive concept, this application also proposes a handheld rapid testing tool, which employs an active detection circuit for the load-carrying capacity of power distribution networks based on a load dynamic model as described in the first aspect.

[0042] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0043] This application proposes an active detection method for the load-carrying capacity of power distribution networks based on a dynamic load model. By retrieving and fusing characteristic quantities of different energy storage units to complete the initial strategy configuration, the entire detection process—including data acquisition, model building, and load loading—is precisely matched with the inherent operating characteristics of energy storage units such as lead-acid batteries, lithium batteries, and supercapacitors. This avoids the accuracy deviation problems caused by generalized detection from the source of detection and adapts to the detection needs of different types of energy storage units. Based on the initial strategy configuration, power supply records and real-time operating parameters are collected to construct a programmable dynamic load model adapted to the type of energy storage unit. This accurately reproduces the actual load characteristics of the power distribution network backup power supply and can effectively simulate normal, transient, and switching cycle loads. This provides theoretical support for load-carrying capacity assessment that closely matches real-world operating conditions, solving the pain points of traditional detection methods that lack dedicated load models and the disconnect between simulation and reality.

[0044] The programmable dynamic load model constructed in this application can effectively achieve high-precision dynamic evaluation of load capacity, allowing load simulation and active detection to accurately match the actual load capacity of the power supply. This ensures the accuracy of load detection and also achieves safe control of the detection process through full-process strategy constraints, avoiding damage to the backup power supply caused by blind loading. Attached Figure Description

[0045] Figure 1 This is a flowchart illustrating the active detection method for the load-carrying capacity of power distribution networks based on a load dynamic model, as shown in an embodiment of the present invention.

[0046] Figure 2 This is a block diagram of the device structure according to an embodiment of the present invention.

[0047] Figure 3 This is a functional structure block diagram of the data acquisition device according to an embodiment of the present invention.

[0048] Figure 4 This is a block diagram of the functional structure of the arithmetic processor according to an embodiment of the present invention.

[0049] Figure 5 This is a flowchart of the data processing program unit in an embodiment of the present invention.

[0050] Figure 6 This is a flowchart of the dynamic load model algorithm processing procedure according to an embodiment of the present invention.

[0051] Figure 7 This is a flowchart of the load capacity assessment algorithm according to an embodiment of the present invention.

[0052] Figure 8 This is a flowchart of the human-machine interface control program according to an embodiment of the present invention. Detailed Implementation

[0053] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0054] Example 1: As Figure 1 As shown, this invention proposes an active detection method for the load-carrying capacity of power distribution networks based on a load dynamic model, comprising:

[0055] Step 1: Set the energy storage unit type of the distribution network backup power supply and retrieve the characteristic quantities of the corresponding energy storage unit type; complete the initial strategy configuration for the load capacity detection of the distribution network backup power supply based on the characteristic quantities;

[0056] Preferably, the characteristic quantities include the charge and discharge characteristics, internal resistance variation law, impact load tolerance threshold, capacity decay characteristics, and constant current output characteristics of each type of energy storage unit.

[0057] In this embodiment, the processing unit of the power distribution network backup power load capacity detection device is equipped with a domestic AT32F chip, which has a built-in FPU floating-point unit and DSP digital signal processor. It has a pre-stored database of characteristic quantities of three types of energy storage units: lead-acid batteries, lithium batteries, and supercapacitors. At the same time, it has a solidified basic dynamic load model construction algorithm, load capacity assessment algorithm, and load array control algorithm, providing a hardware and algorithm foundation for the initial strategy configuration.

[0058] In this embodiment, the specific implementation process of step 1 includes:

[0059] Step 1.1: Setting the energy storage unit type and retrieving characteristic quantities;

[0060] The testing personnel select lithium battery as the energy storage unit type of the backup power supply of the power distribution network under test through the human-machine interface of the testing device. After receiving the type selection instruction, the processing unit of the testing device automatically retrieves the corresponding characteristic quantities of the lithium battery from the characteristic quantity database. The characteristic quantities specifically include the charging and discharging characteristics of the lithium battery, such as nominal charging and discharging voltage 3.2V / cell, charging and discharging rate 1C, internal resistance variation law, such as normal operating internal resistance ≤50mΩ, internal resistance increases linearly with capacity decay, impact load tolerance threshold, preferably instantaneous impact current ≤8C, continuous impact duration ≤200ms, capacity decay characteristics, preferably capacity decay ≤20% after 500 charge and discharge cycles, constant current output characteristics, preferably rated constant current output current 5A, constant current output voltage fluctuation ≤±0.5%.

[0061] Step 1.2: Extract core parameters of features to form a structured parameter set;

[0062] The processing unit performs structured analysis on the retrieved lithium battery features, extracts the core quantification parameters of each feature, removes redundant information, and forms a structured parameter set of lithium battery features. The parameter set is stored in the temporary cache of the processing unit in the form of key-value pairs. Specifically, it includes: core charging and discharging parameters, internal resistance threshold parameters, impact load tolerance parameters, capacity decay parameters, and constant current output parameters, providing accurate and standardized parameter basis for subsequent algorithm optimization and strategy configuration.

[0063] Step 1.3: Feature fusion and targeted correction of the core detection algorithm;

[0064] The processing unit integrates the structured parameter set of lithium battery features into the pre-stored core detection algorithm, and performs targeted parameter and logic corrections on the dynamic load model construction algorithm, load capacity assessment algorithm, and load array control algorithm to ensure that the computational logic of each algorithm is precisely adapted to the inherent operating characteristics of the lithium battery. The specific correction method is as follows:

[0065] For the dynamic load model construction algorithm: adjust the parameter identification logic, increase the identification weight of transient load current and voltage fluctuations, and adapt to the characteristics of lithium batteries being sensitive to transient impact loads; at the same time, correct the load characteristic fitting threshold, limit the current fitting range of the model to 0~8C, and match the impact load tolerance threshold of lithium batteries.

[0066] For the load capacity assessment algorithm: the load assessment weights are optimized by setting the assessment weights for internal resistance change and instantaneous current carrying time to 30% and 35% respectively, weakening the assessment weight of static voltage parameters to avoid assessment bias caused by a single static parameter; at the same time, a load capacity judgment threshold specific to lithium batteries is set, such as directly judging the impact load capacity as not meeting the standard when the instantaneous impact current exceeds 8C.

[0067] For the load array control algorithm: Adjust the drive threshold boundary, set the maximum drive current threshold of the load array to 8C, and set the drive voltage fluctuation threshold to ±0.5%, strictly match the constant current output and impact load tolerance characteristics of the lithium battery, and avoid lithium battery damage caused by over-threshold drive from the algorithm bottom layer.

[0068] The modified detection core algorithm is deeply adapted to the characteristics of lithium batteries and stored in the algorithm storage area of ​​the computing processor, serving as the basis for subsequent modeling, evaluation, and load control algorithms.

[0069] Step 1.4: Configure differentiated voltage and current synchronous acquisition strategy;

[0070] Preferably, the processing unit is configured with a customized voltage and current synchronous acquisition strategy based on the parameter sensitivity reflected by the structured parameter set of lithium battery characteristic quantities (lithium battery voltage and current are sensitive to transient changes and have low tolerance to high-frequency interference signals). This strategy specifies the A / D sampling accuracy, sampling rate, filtering method, signal transformation rules, and amplitude limiting protection threshold. Specific configuration parameters are as follows:

[0071] 1. The A / D sampling accuracy is set to 16 bits and the sampling rate is set to 2MSPS to adapt to the rapid changes in transient parameters of lithium batteries and ensure the accuracy of the collected data;

[0072] 2. The filtering method adopts a second-order low-pass filter with a cutoff frequency set to 1kHz, which effectively filters out high-frequency interference noise in voltage and current signals while retaining the effective characteristics of transient changes.

[0073] 3. Signal conversion rules: A combination of linear resistor voltage divider + linear opto-isolation + operational amplification is used to linearly convert the actual output voltage of the lithium battery (multiple cells connected in series to 32V) into a standard linear voltage of 0~3.3V, matching the signal receiving threshold of the operational processor;

[0074] 4. Limiting protection threshold: The voltage limit is set to 3.3V±0.05V, and the current sampling signal limit is set to the electrical signal amplitude corresponding to the 8C inrush current to prevent overvoltage and overcurrent spike signals from damaging the acquisition circuit and the processing unit.

[0075] The aforementioned differentiated data acquisition strategy is embedded in the data acquisition control module of the detection device, directly guiding the subsequent acquisition process of ledger information and operating parameters.

[0076] Step 1.5: Safe pre-setting of basic load loading parameters;

[0077] Preferably, the processing unit presets the basic parameters of load loading based on the impact load tolerance threshold and constant current output characteristics of the lithium battery. This includes specifying the initial load amplitude, single load duration, number of load steps, load interval, and the frequency of the switch opening and closing cyclic loading simulation. All initial load parameters are strictly controlled within the safety tolerance boundary of the lithium battery, while also meeting the requirements of the power regulations for simulating three switching operations. The specific preset parameters are as follows:

[0078] Initial load amplitude: The normal load amplitude is set to 5A, and the transient impact load amplitude is set to 40A;

[0079] Single loading duration: The single loading duration for normal load is set to 5s, and the single loading duration for transient impact load is set to 200ms, which does not exceed the continuous withstand time of the lithium battery impact load.

[0080] Loading steps and intervals: Loading is completed in 3 steps to match the requirements of three switching operations. The normal load loading interval is set to 1 second, and the transient impact load loading interval is set to 5 seconds to ensure that the lithium battery has enough recovery time.

[0081] The simulated frequency for the cyclic loading of switch opening and closing is set to 50Hz to match the frequency characteristics of actual switch operations in the distribution network.

[0082] The basic parameters for load loading are stored in the load control module of the computing processor, serving as the initial execution parameters for subsequent load array loading.

[0083] Step 1.6: Setting up a multi-level safety protection strategy for the detection process;

[0084] Preferably, the processing unit, based on the capacity decay characteristics and overcurrent and overvoltage tolerance thresholds of lithium batteries, sets a multi-level safety protection strategy during the testing process, clearly defining the load derating ratio, shutdown triggering conditions, and emergency handling rules under abnormal operating conditions, thereby achieving proactive safety control of the testing process. The specific strategy is as follows:

[0085] Level 1 warning strategy: When the internal resistance of the lithium battery is detected to be ≥40mΩ, which is close to the 50mΩ threshold, and the instantaneous load current reaches 7C, a level 1 warning is triggered. The processing unit automatically reduces the load amplitude by 20% and displays a warning prompt on the human-machine interface.

[0086] Level 2 shutdown strategy: When any of the following conditions are detected, such as lithium battery internal resistance ≥50mΩ, instantaneous load current ≥8C, or voltage fluctuation exceeding ±0.5%, a level 2 shutdown is triggered. The processing unit immediately sends a command to stop the load loading and disconnects the electrical connection between the detection device and the lithium battery, while recording the shutdown fault parameters.

[0087] Capacity decay correlation strategy: When the lithium battery capacity decay is detected to be ≥15%, that is, close to the 20% threshold, the loading parameters are automatically adjusted to reduce the number of transient impact loads from 3 to 2, so as to avoid secondary damage to the decaying battery.

[0088] The aforementioned multi-level safety protection strategy is embedded into the logic judgment program unit of the detection device, enabling real-time safety monitoring and automatic emergency handling during the detection process.

[0089] Step 1.7: Initial strategy set integration and solidification;

[0090] The processing unit integrates the aforementioned feature-optimized core detection algorithm, customized voltage and current synchronous acquisition strategy, preset load loading basic parameters, and set multi-level safety protection strategies to form a unified initial strategy set for lithium battery power grid backup power detection. This strategy set is then programmed into the processing unit of the detection device, serving as the sole basis for subsequent power grid backup power ledger information collection, programmable dynamic load model construction, load capacity assessment, and load array optimization and control. This ensures that all subsequent detection steps follow this strategy set, achieving standardization, systematization, and controllability of the detection process.

[0091] This embodiment completes the initial strategy configuration of the lithium battery energy storage unit through the above steps. If the energy storage unit under test is a lead-acid battery or a supercapacitor, it is only necessary to select the corresponding type in step 1.1, retrieve its characteristic quantities, and then follow the same process as this embodiment to complete the algorithm correction, acquisition strategy configuration, loading parameter preset and safety strategy setting according to the core parameters of its characteristic quantities. This will form a unique initial strategy set for the corresponding energy storage unit. The operation process is unified and highly adaptable, which greatly improves the adaptability of the detection device to different types of energy storage units.

[0092] Step 2: Based on the initial strategy configuration, collect the ledger information and voltage and current operating parameters of the backup power supply in the distribution network, and construct a programmable dynamic load model adapted to this type of energy storage unit; specifically:

[0093] Step 2.1: Collect basic ledger information and real-time operating parameters based on the initial strategy set;

[0094] After the detection device is started, the computing processor automatically calls the voltage and current synchronous acquisition strategy customized in the initial strategy set, controls the data acquisition module of the detection device, and synchronously completes the acquisition of basic ledger information and real-time operating parameters of the power distribution network backup power (lithium battery), ensuring that the acquired data dimensions are complete and the accuracy meets the standards, and avoiding the distortion of subsequent model construction due to missing data dimensions or signal distortion.

[0095] Preferably, the basic ledger information is collected as follows: the testing personnel manually enter the basic ledger information of the backup power supply of the distribution network under test through the human-computer interaction interface, or connect with the operation and maintenance management system of the backup power supply of the distribution network through the serial port interface of the testing device to automatically read the ledger data. The ledger information is collected strictly in accordance with the requirements of the initial strategy set, specifically including: the rated voltage of the backup power supply of the distribution network is 32V (10 lithium batteries in series), the rated current is 5A, the rated capacity is 100Ah, the operation and maintenance records (32 charge and discharge times in the past 6 months, the last maintenance time is 1 month ago), and the historical load characteristics (normal load current is 3~4A, the average number of impact loads per month is 5, and the peak current of the impact load is 38A). All ledger information is stored in a structured format in the storage unit of the computing processor as the basic reference data for model construction.

[0096] 2. Real-time Parameter Acquisition: The high-precision voltage and current sensors of the detection device, according to the acquisition parameters set in the initial strategy set, including 16-bit A / D sampling accuracy, 2MSPS sampling rate, second-order low-pass filtering, 0~3.3V signal transformation, and voltage limiting of 3.3V±0.05V, simultaneously acquire the real-time voltage and current analog signals of the lithium battery. During the acquisition process, the sensor transmits the analog signal to the data acquisition module, which sequentially performs signal transformation, low-pass filtering, and limiting protection processing. Specifically, through linear resistor voltage division, linear opto-isolation, and operational amplification, the actual 32V output voltage of the lithium battery is transformed into a standard linear voltage of 0~3.3V, which is then converted into a digital quantity by a 16-bit A / D converter, ultimately forming a structured real-time detection data source, which is transmitted to the temporary buffer of the processing unit for subsequent preprocessing.

[0097] Step 2.2: Data source preprocessing and modeling algorithm, basic model invocation;

[0098] The processing unit first retrieves the dynamic load model construction algorithm optimized by lithium battery features from the initial strategy set, which is the algorithm modified in step 1.3. At the same time, it loads the preset polynomial dynamic load basic model pre-installed in the processing unit. This basic model is a general mathematical template that includes linear terms, quadratic terms and higher-order polynomial terms, which can adapt to the modeling requirements of different types of energy storage units. Its initial parameters have been preliminarily adapted according to the characteristics of lithium batteries.

[0099] Subsequently, the processing unit creates a time-series data cache array, with a preferred cache capacity of 1024 data sets. The structured ledger information collected in step 2.1 is used to extract historical load characteristic parameters and preprocessed real-time detection data sources, which are then synchronously imported into the time-series data cache array. To further eliminate high-frequency interference and invalid data and ensure the reliability of the modeling data, the processing unit preprocesses the data in the cache array using a sliding window mean filtering algorithm. The sliding window size is set to 16, and the mean of the data within the window is taken as valid data, while instantaneous spike interference data is filtered out. After preprocessing, a standardized modeling data source is formed for subsequent model iteration and solution.

[0100] Preferably, before importing the aforementioned data source into the cache array, the method further includes:

[0101] The acquired voltage and current analog signals are sequentially processed by signal transformation, low-pass filtering, and amplitude limiting protection. The backup power supply voltage is transformed into a standard linear voltage of 0~3.3V through linear resistor voltage division, linear opto-isolation, and operational amplification. Then, it is converted into a digital quantity by A / D conversion with the accuracy set in the initial strategy configuration, forming a structured detection data source.

[0102] Step 2.3: Iteratively solve for the optimal load characteristic parameters based on the RLS identification algorithm;

[0103] The processing unit reads the load characteristic parameters of the lithium battery under the historical stable operating conditions from the structured ledger information, including the normal load current of 3~4A, the peak current of the impact load of 38A, and the load change rate of 0.5A / s. It then merges these historical parameters with the real-time modeling data source after preprocessing in step 2.2 as the input data for the RLS identification algorithm.

[0104] According to the requirements of the initial strategy set, the recursive least squares (RLS) identification algorithm is used to iteratively solve the linear model part in the preset polynomial dynamic load basic model. During the iteration process, the computing processor sets the number of iterations to match the lithium battery type. Combining the charging and discharging characteristics and internal resistance variation law of the lithium battery, the number of iterations is set to 100, and the iteration convergence threshold is set to 0.001 to ensure the accuracy of the iteration results.

[0105] Preferably, during the iterative solution process, the RLS identification algorithm continuously adjusts the linear term coefficients of the polynomial dynamic load basic model. Combining the transient characteristics of lithium batteries, it focuses on optimizing the parameter identification accuracy under impact load conditions. Finally, after 100 iterations, it converges to obtain the optimal load characteristic parameters that are suitable for the backup power supply of the lithium battery distribution network under test. In this embodiment, the optimal load characteristic parameters specifically include: load dynamic impedance change coefficient 0.02Ω / A, normal load fitting coefficient 0.98, transient impact load response coefficient 1.03, and switch opening and closing load cycle attenuation coefficient 0.01. These optimal load characteristic parameters fully reflect the actual load characteristics of the current power supply under test and provide core parameter support for subsequent model construction.

[0106] Step 2.4: Calculate dynamic load characteristics and dynamically correct the model to obtain the final programmable dynamic load model;

[0107] Calculate the change in dynamic load impedance and load change rate of the backup power supply in the distribution network over time. The dynamic load impedance reflects the impedance change pattern of the load under normal and transient conditions, while the load change rate reflects the dynamic rate of change of the load. Specific calculations include:

[0108] The load impedance changes over time, calculated at 10ms intervals, to adapt to the transient load change rate of lithium batteries.

[0109] The load change rate, combined with real-time current data, calculates the change in load current every 10ms.

[0110] Simultaneously, load characteristic curves are calculated for three operating conditions: normal load, transient impact load, and cyclic load of switch opening and closing, to ensure that the model can cover various load scenarios in the actual operation of the distribution network.

[0111] To further improve the model's adaptability, the computing processor incorporates the characteristic quantities of the lithium battery again, and performs targeted dynamic corrections to the model: combining the lithium battery's charge and discharge voltage of 3.2V / cell and the 8C impact load tolerance threshold, the load simulation amplitude boundary of the model is corrected; combining the linear increase of the lithium battery's internal resistance with capacity decay, the load impedance change logic of the model is corrected; combining the characteristic of the lithium battery's constant current output voltage fluctuation ≤±0.5%, the constant current output simulation accuracy of the model is optimized, so that the model's computing logic is fully adapted to the inherent operating characteristics of the lithium battery under test.

[0112] After correction, the final programmable dynamic load model is obtained. This model is stored in the model storage area of ​​the computing processor and has the following core characteristics: it can accurately reflect the load target value of the current distribution network backup power supply under test, which is 5A under normal conditions and 40A under transient conditions; and dynamic automatic constant current value; it can effectively simulate normal load, transient impact load, and cyclic load of switch opening and closing under actual distribution network conditions, and strictly meet the simulation requirements of three switching operations required by power regulations; at the same time, it can synchronously collect, identify, and judge the characteristic quantity changes of the distribution network backup power supply during various load loading processes, providing a realistic, accurate and reliable theoretical model support for subsequent load capacity assessment.

[0113] This embodiment completes the construction of a programmable dynamic load model adapted to lithium battery energy storage units through the above steps. The entire process strictly follows the requirements of the initial strategy set. The constructed model not only conforms to the inherent characteristics of lithium batteries but also matches the individual operating parameters of the backup power supply of the current distribution network under test. This effectively solves the problem of the disconnect between traditional general load models and actual power supply characteristics, ensuring the accuracy and reliability of subsequent load capacity assessment. If the energy storage unit under test is a lead-acid battery or a supercapacitor, it is only necessary to collect the corresponding ledger information and operating parameters, call its dedicated optimized dynamic load model construction algorithm, set the RLS iteration number to match its characteristics, incorporate its feature quantities to complete the model correction, and thus construct the corresponding type of programmable dynamic load model.

[0114] Step 3: Input the real-time collected operating parameters such as voltage and current of the distribution network backup power supply into the programmable dynamic load model to evaluate the load-carrying capacity and output a dynamic adjustment strategy for the load-carrying capacity.

[0115] It should be noted that the load capacity assessment is specifically divided into three core dimensions, including:

[0116] 1. Assessment of normal load carrying capacity: The real-time collected normal load current (3~4A) is compared with the normal load characteristic curve simulated by the model to evaluate the voltage stability and current output consistency of the lithium battery under continuous normal load, and to determine whether it can maintain the rated constant current output (5A) for a long time. At the same time, the capacity decay impact under normal load is evaluated by combining the internal resistance change data.

[0117] 2. Transient impact load carrying capacity assessment: Simulate distribution network switch operation and sudden fault scenarios, input the real-time collected transient impact current (up to 38~40A) into the model, evaluate the voltage drop amplitude and current response speed of lithium battery under 200ms instantaneous impact load, determine whether it meets the 8C impact load tolerance threshold requirement, and record the voltage and internal resistance recovery after a single impact.

[0118] 3. Evaluation of the load-carrying capacity of the switch opening and closing cycle load: In accordance with the three-stage switch operation specifications required by the power regulations, the model simulates three cycles of impact load (40A, 200ms / cycle, 5s interval) to evaluate the load-carrying stability and performance degradation of the lithium battery after multiple impacts, determine the number of impacts it can withstand, and ensure that the evaluation results meet the requirements of the distribution network emergency power supply.

[0119] During the evaluation process, the computing processor synchronously integrates the real-time temperature data and internal resistance change data of the lithium battery, and combines them with the operation and maintenance records and historical load characteristics in the ledger to dynamically correct the evaluation results, avoiding misjudgments caused by a single parameter or instantaneous data, and ensuring the scientific and accurate nature of the evaluation results.

[0120] Furthermore, based on the multi-dimensional load capacity assessment results, the processing unit automatically outputs a dynamic adjustment strategy for the load capacity of the backup power supply of the lithium battery distribution network under test. This strategy strictly combines the safety protection requirements of the initial strategy set and the boundary of lithium battery characteristic quantities, while also adapting to the subsequent load array optimization and active loading detection requirements. Specifically, it includes three major categories of strategies, with clearly defined executable parameters and rules:

[0121] 1. Dynamic optimization strategy for loading parameters: Based on the normal load assessment results, if the lithium battery voltage fluctuation exceeds ±0.5%, the subsequent normal load amplitude will be adjusted to 4.5A (derating by 10%); based on the transient impact assessment results, if the voltage drop exceeds 1V after a single impact, the subsequent transient impact load amplitude will be adjusted to 36A (90% of the 8C threshold), and the loading interval will be extended from 5s to 8s to ensure that the lithium battery has sufficient recovery time.

[0122] 2. Load array configuration adjustment strategy: Based on the actual load capacity of the lithium battery obtained from the evaluation, output load array combination adjustment instructions are given to clarify the number of IGBT power transistors in the linear load array in series and parallel, so as to ensure that the output load of the load array is accurately matched with the actual load capacity of the lithium battery and avoid detection deviation caused by excessive or insufficient load.

[0123] 3. Abnormal Operating Condition Safety Handling Strategy: Clearly define the judgment criteria and emergency handling rules for abnormal operating conditions. If any of the following conditions are detected during the assessment: lithium battery internal resistance ≥45mΩ, instantaneous inrush current ≥40A, or voltage fluctuation exceeding ±0.8%, immediately trigger a 30% load derating command. If a voltage drop exceeding 2V or a temperature exceeding 60℃ is detected, immediately trigger a load suspension command, simultaneously display abnormal warning information on the human-machine interface, record abnormal parameters, and ensure testing safety and lithium battery operation safety.

[0124] The aforementioned dynamic load capacity adjustment strategy is stored in the processing unit in the form of structured instructions and synchronously transmitted to the load drive module and the human-machine interface.

[0125] Step 4: Optimize the load array by combining the energy storage unit type, initial strategy configuration and dynamic load capacity adjustment strategy, perform active load detection on the optimized backup power supply, generate the final evaluation report of the load capacity of the distribution network backup power supply, and visualize it on the human-machine interface.

[0126] In this embodiment, load array optimization is performed, specifically including:

[0127] First, extract three key pieces of evidence, including:

[0128] First, the characteristic quantity boundaries of the lithium battery energy storage unit, specifically the impact load tolerance threshold (instantaneous impact current ≤8C, continuous impact duration ≤200ms) and constant current output characteristics (rated constant current output current 5A, voltage fluctuation ≤±0.5%).

[0129] Second, the load array control algorithm is fixed in the initial strategy configuration (the algorithm optimized by lithium battery characteristics in S103 above). This algorithm has preset the load array driving logic and combination rules adapted to lithium batteries.

[0130] Third, the loading parameter requirements in the dynamic load capacity adjustment strategy output in step 3 are as follows: normal loading amplitude of 4.5A, transient impact loading amplitude of 36A, loading interval extended to 8s, and load array configuration adjustment instructions.

[0131] In this embodiment, the linear load array consists of several low-switching-loss, high-speed IGBT power transistors (preferably IRF3205), a manganese-copper shunt, and a current feedback circuit. All IGBT power transistors are connected in parallel with fast recovery full-current diodes, allowing for flexible load adjustment through series-parallel combinations. Based on the above three criteria, the processing unit outputs combined control commands through the load array control algorithm to adjust the number and connection method of the IGBT power transistors in series and parallel.

[0132] Normal load condition: Matching a normal load amplitude of 4.5A, controlling 6 groups of IGBT power transistors connected in parallel to ensure that the normal load current output of the load array is stable at around 4.5A, and the voltage fluctuation meets the requirement of ±0.5%;

[0133] Transient impact load condition: Matching a transient impact load amplitude of 36A, controlling 36 groups of IGBT power transistors connected in parallel (divided into 3 groups, 12 in each group), while adjusting the conduction rate of the connection lines to ensure that the instantaneous output of 36A impact current is not more than 200ms.

[0134] Switching on / off cyclic loading condition: Combining the three-stage switching operation specifications, the IGBT power transistor's on and off timing is controlled, and an 8-second loading interval is matched to achieve precise control of three-stage cyclic impact loading. At the same time, a manganese-copper shunt is connected in series, and the load current is fed back in real time through micro-voltage acquisition, providing data support for the dynamic adjustment of the load array.

[0135] Furthermore, after the load array combination optimization is completed, the computing processor outputs a test signal through the load drive module to detect the actual output load parameters of the load array, confirming that its normal load output accuracy error is ≤±3% and the transient impact load response delay is ≤10ms, which meets the requirements of the initial strategy configuration and dynamic adjustment strategy, thus completing the load array optimization. The optimized load array is stored in the load control module, waiting for subsequent active loading detection instructions.

[0136] After the load array is optimized, the processing unit initiates the active loading detection process. This process uses a non-stop detection mode, requiring no changes to the existing application mode of the distribution network. It directly drives the optimized linear load array to perform active loading detection on the lithium battery backup power supply under test, achieving dynamic closed-loop control throughout. The specific process is as follows:

[0137] 1. Drive Signal Conversion and Closed-Loop Control Construction: The arithmetic processor outputs a 0~3.3VD / A low-voltage control signal that matches the optimized load array and transmits it to the load drive circuit module. The load drive module amplifies the low-voltage control signal into a 0~12V drive signal through an amplification circuit composed of operational amplifiers. At the same time, it collects the real-time current feedback signal of the linear load array through a differential feedback circuit. The drive signal and the current feedback signal are compared in real time by a comparison circuit to form a dynamic closed-loop control circuit. This ensures that the output load current of the load array is accurately matched with the preset loading parameters and synchronously follows the load state changes of the lithium battery.

[0138] 2. Multi-condition active loading test execution: In accordance with the dynamic adjustment strategy for load capacity and the requirements of power regulations, active loading tests are performed sequentially under three operating conditions, maintaining a continuous operation without shutting down the system. Specific test procedures are as follows:

[0139] Normal load test: Drive the optimized load array to continuously load at an amplitude of 4.5A for 5 minutes to simulate the normal operation load of the distribution network. Simultaneously collect the real-time response parameters of the lithium battery, such as voltage (target 32V±0.5%), current (target 4.5A), and temperature (normal operating temperature 0~55℃), as well as the actual output load current of the load array, and record the load stability under normal load.

[0140] Transient impact load detection: Drive the load array to perform a single transient impact load with an amplitude of 36A and a duration of 200ms to simulate the operation scenario of the distribution network switch. Simultaneously collect the voltage drop amplitude and current response speed of the lithium battery during the impact, as well as the voltage and internal resistance recovery after the impact, and record the load performance of a single impact.

[0141] Switch opening and closing cyclic loading test: In accordance with the three-stage switch operation specifications, the load array is driven to perform three cyclic impact loadings with an amplitude of 36A and a duration of 200ms, with a loading interval of 8s. This simulates multiple switch operation scenarios after a sudden power grid fault. The lithium battery response parameters after each impact are collected synchronously to evaluate the number of impacts it can withstand and the degree of performance degradation.

[0142] 3. Safety Management During Testing: During the loading and testing process, the processing unit compares the collected lithium battery response parameters with the multi-level safety protection thresholds set in the initial strategy configuration in real time, strictly implementing the anomaly handling rules in the dynamic load capacity adjustment strategy to ensure testing safety.

[0143] If the internal resistance of the lithium battery is detected to be ≥45mΩ, the instantaneous inrush current is ≥38A, or the voltage fluctuation exceeds the ±0.8% level 1 warning threshold, immediately execute the load derating operation of 30% according to the dynamic adjustment strategy, adjust the normal load amplitude to 3.15A and the transient inrush load amplitude to 25.2A, and at the same time display the warning prompt on the human-machine interface and record the warning parameters;

[0144] If a sudden drop in lithium battery voltage exceeding 2V, temperature exceeding 60℃, or internal resistance ≥50mΩ secondary termination threshold is detected, a load termination command is immediately triggered to stop active load detection, disconnect the electrical connection between the load array and the lithium battery, generate abnormal warning information, and record the time of abnormality and abnormal parameters in detail to facilitate subsequent fault investigation.

[0145] In this embodiment, the active loading detection did not trigger an abnormal termination command throughout the entire process. Only after the second cycle of impact loading was a lithium battery voltage fluctuation of ±0.7% detected, triggering a first-level warning. After the load derating operation was executed, the detection continued normally.

[0146] After the active loading detection is completed, the processing unit summarizes all the detection data, combines the load capacity assessment results from step 3 with the simulation data from the programmable dynamic load model, and generates a final assessment report on the load capacity of the distribution network backup power supply.

[0147] Preferably, the final evaluation report should include at least the following core contents:

[0148] 1. Basic testing information: Type of backup power storage unit of the distribution network under test, rated parameters, testing time, and testing conditions;

[0149] 2. Load capacity assessment results: load capacity under normal load, load capacity under transient impact load, and load capacity under cyclic load of switch opening and closing;

[0150] 3. Health status assessment: Based on the test data, the health status of the lithium battery is determined to be good, with an internal resistance of 43mΩ (≤50mΩ), a capacity decay rate of 12%, and no significant performance degradation.

[0151] 4. Abnormal Situation Record: This test only triggered one Level 1 warning. The reason for the warning, the handling process, and the effect of the handling are recorded.

[0152] 5. Operation and maintenance recommendations: Based on the assessment results, it is recommended to conduct a routine load test every 3 months and a cyclic impact load test every 6 months, and regularly monitor changes in internal resistance to ensure the stability of the lithium battery's load-bearing capacity.

[0153] After the final evaluation report is generated, the processing unit transmits the report data, real-time operating parameters during the testing process, intermediate results of load capacity evaluation, dynamic adjustment strategies, and abnormal warning information to the serial port human-machine interface with a touch screen through the TTL driver interface circuit, so as to realize the visualization display.

[0154] In Example 2, this application also proposes an active detection circuit for the load-carrying capacity of power distribution networks based on a load dynamic model. The circuit adopts an active detection method for the load-carrying capacity of power distribution networks based on a load dynamic model as described in Example 1.

[0155] Preferably, the active detection circuit for the load-carrying capacity of the distribution network power supply based on the load dynamic model includes:

[0156] Data acquisition circuit module: includes signal conversion circuit, low-pass filter circuit and amplitude limiting protection circuit; used for synchronous acquisition, conversion, filtering and protection of voltage and current signals, and outputting digital signals to the arithmetic processing circuit module;

[0157] The arithmetic processing circuit module includes peripheral basic circuits, A / D conversion circuits, chip main control circuits, and DAC interface driver circuits; the chip main control circuit has built-in data processing programs, dynamic load model algorithm programs, load capacity assessment algorithm programs, and logic control programs.

[0158] The load drive circuit module includes an amplifier circuit, a differential feedback circuit, and a comparator circuit. The amplifier circuit is composed of an operational amplifier, which amplifies the 0~3.3V low voltage signal input from the operational processing circuit module into a 0~12V drive signal. The differential feedback circuit acquires the real-time current signal of the linear load array circuit module and feeds it back to the inverting input of the comparator circuit. The comparator circuit compares the amplified drive signal with the current feedback signal in real time to form a dynamic closed-loop control circuit and outputs the drive signal to the linear load array circuit module.

[0159] The linear load array circuit module consists of several low-switching-loss high-speed IGBT power transistors, a manganese-copper shunt, and a current feedback circuit. Each IGBT power transistor is connected in parallel with a fast recovery full-current diode. Multiple IGBT power transistors are connected in series and parallel to form a programmable linear load array. The load current is linearly regulated by the gate signal. The manganese-copper shunt is connected in series with the load array and uses micro-voltage acquisition to achieve real-time load current detection. The current signal is then transmitted to the differential feedback circuit of the load drive circuit module via the current feedback circuit.

[0160] The human-computer interaction circuit module includes a TTL driver interface circuit and a touch serial port display circuit. The TTL driver interface circuit enables bidirectional data interaction with the arithmetic processing circuit module, receiving detection data, evaluation results, and control commands and transmitting them to the touch serial port display circuit. The touch serial port display circuit is a serial port display circuit with a touch screen, which completes the functions of setting detection parameters, real-time display of detection status, visualization of load capacity evaluation results, and issuing manual commands.

[0161] Preferably, the detection circuit further includes:

[0162] The AT32F chip in the computing circuit module also integrates a logic judgment program unit and a human-machine interface control program unit. The logic judgment program unit completes the threshold judgment of the running parameters and the logic triggering of abnormal working conditions during the detection process. The human-machine interface control program unit realizes the instruction parsing, status feedback and dynamic interface update of the touch serial port display circuit.

[0163] In embodiment three, this application also proposes a handheld rapid testing tool, which employs an active detection circuit for the load-carrying capacity of power distribution networks based on a load dynamic model as described in the first aspect.

[0164] In this embodiment, as Figure 2-3 As shown, the handheld rapid testing tool includes a data acquisition unit 1, a processing unit 2, a load drive unit 3, a linear load array 4, and a human-computer interaction unit 5.

[0165] As a preferred embodiment, the data acquisition unit 1, whose core function is to transform the electrical quantity to be acquired, includes a signal conversion circuit 11, a filtering circuit 12, and a protection circuit 13. It synchronously acquires and feeds back voltage and current during the detection process, converting the voltage and current signals of the test object into digital quantities and sending them to the processing unit as input factors for the algorithm for data processing and judgment. Specifically, the signal conversion circuit 11 uses a combination of linear resistor voltage division, linear opto-isolation, and operational amplification to convert the backup power supply voltage into a linearly related voltage of 0~3.3V. The filtering circuit 12 uses low-pass filtering technology to filter the input signal before sending it to the subsequent circuit for acquisition. The protection circuit 13 uses dual fast diodes to prevent the output voltage amplitude from being too large and thus limiting it. As a preferred embodiment, the processing unit 2 is the core part of the device. Its core function is to perform data processing, calculation algorithms, logical judgments, and control software operation functions. It includes peripheral basic circuits 21, an A / D converter 22, a CPU main control chip 23, and an interface driver circuit 24 to realize the functions of the device.

[0166] The peripheral basic circuit 21 mainly consists of peripheral circuits, including the design of an 8MHz crystal resonator and an NRST reset circuit. The 8MHz crystal resonator can provide a high-precision clock source for the system. The design of the NRST reset circuit, combined with WDT reset and WWDT reset, realizes software reset and low-power management reset functions, achieving the reset purpose of events such as power-on reset, low-voltage reset, or returning from standby mode.

[0167] The A / D converter 22 uses a peripheral device to convert analog input signals into 12-bit digital signals, with a maximum sampling rate of 2MSPS. Different acquisition channels have their own independent trigger detection circuits, and the sampling time can be configured independently. The conversion sequence management supports a variety of different multi-channel conversion controls.

[0168] The CPU main control chip 23 adopts the domestically produced AT32F chip solution, which integrates high efficiency, low power consumption and reliability. It has a built-in single precision floating-point unit (FPU) and digital signal processor (DSP), and is equipped with rich peripherals and flexible clock control mechanism to realize a low-power processor chip configuration design with low gate count, low interrupt latency and low cost debugging. The interface driver circuit 24 uses a 12-bit digital input digital-to-analog converter (DAC) design to realize an analog voltage output signal between 0 and 3.3V, which is used to control the load driving device in the subsequent stage.

[0169] The CPU main control chip 23 includes programs such as an execution data processing program unit 231, a dynamic load model algorithm processing program unit 232, a load capacity assessment algorithm program unit 233, and a human-machine interface control program unit 234.

[0170] The execution flow of the data processing unit 231 includes: hardware initialization to complete the device configuration before acquisition; initializing the ADC (analog-to-digital converter) to set the sampling accuracy (e.g., 12-bit / 16-bit) and sampling frequency (e.g., 1kHz); configuring the signal conditioning circuit of the voltage / current sensor; initializing the communication interface (optional); and acquiring the raw voltage / current signal. After setting the core parameters for acquisition and calculation, including the sampling period (e.g., 100ms / time), acquisition channels, and calibration parameters, the ADC sampling is started to acquire the analog signals of the voltage and current channels respectively, convert them into digital quantities, and store them in a temporary buffer array. The next step is to perform noise and error elimination and mean filtering to remove high-frequency interference, ensuring data validity before further processing, data storage and output, and exception handling to ensure device safety and program stability.

[0171] The dynamic load model algorithm processing unit 232 updates load characteristic parameters in real time based on time-series data. It simulates dynamic output load by adjusting for voltage, real-time load feedback, or time. Its flowchart needs to cover the entire chain from data input, dynamic parameter identification, characteristic calculation, verification feedback, to loading output requirements, while also reflecting temporality and iterativeness. The following is a detailed flowchart design, including the core steps of identifying the preset dynamic load model using a polynomial dynamic model, a linear model using a recursive least squares (RLS) algorithm, and dynamic load characteristic calculation. The preset dynamic load model uses a polynomial dynamic model, where the dynamic load impedance reflects the change in load impedance over time, and the load change rate is calculated to reflect the dynamic change speed of the load.

[0172] The core of the human-machine interface control program unit 234 is to realize two-way interaction between the user and the device, including the entire process of initialization, command parsing, device status feedback, and dynamic interface updates, while ensuring the real-time nature of the interactive response and the intuitiveness of the operation. The following is a detailed flowchart design, which includes sub-units such as permission verification, exception handling, and resource release.

[0173] As a preferred embodiment, the load drive unit 3's core function is to receive the D / A low-voltage signal output by the processing unit, compare it with the feedback detection signal at the load end to form a drive signal for the linear load array, and drive the load array section. This process includes an amplifier circuit 31, a feedback circuit 32, and a comparator circuit 33. Specifically, the amplifier circuit 31 amplifies the 0~3.3V analog signal from the controller unit to 0~12V through an operational amplifier circuit, improving the driving and adjustment capabilities of the subsequent drive circuit. The feedback circuit 32 uses a differential feedback design to feed the output current signal back to the inverting input of the comparator circuit for synchronously following and adjusting the output load current. The comparator circuit 33 uses the comparison current generated by the operational amplifier to compare the 0~12V drive voltage signal with the output current feedback signal, forming a dynamic closed loop for synchronously following and adjusting the output load current. Ultimately, this controls the subsequent linear load array 4.

[0174] As a preferred embodiment, the linear load array 4 has the core function of forming a programmable linear load combination array. It receives output signals from the load drive unit and control signals from the processing processor 2, and freely combines the linear load array to load the backup power supply of the detected distribution network with a load matching the configuration. It includes linear load power transistors 41 and a current feedback circuit 42. Specifically, the linear load power transistors 41 employ low-switching-loss, high-power IGBT technology and feature a fast-recovery full-current reverse-parallel diode. A control method that linearly changes the current through the gate (G) terminal achieves linear changes in the IGBT's thermal resistance effect, thereby controlling the current. Multiple linear load power transistors 41 are connected in series and parallel to form a load array. The current feedback circuit 42 uses a manganese-copper shunt technology connected in series with the load array to collect micro-voltage data and send the feedback to the load drive unit 3.

[0175] As a preferred embodiment, the human-computer interaction 5, whose core functions are to display data, status, set functions, and perform human-computer operations, achieves intuitive human-computer interaction. It includes a display screen interface 51 and a serial port display screen 52. Specifically, the display screen interface 51 refers to the display screen connecting to the processor via a TTL driver, enabling data interaction and driving, controlling, and processing the functions displayed on the screen. The serial port display screen 52 utilizes touchscreen technology to achieve fast, intuitive, and complete interface display and operation.

[0176] Although exemplary embodiments have been described herein with reference to the accompanying drawings, it should be understood that the above exemplary embodiments are merely illustrative and are not intended to limit the scope of the invention. Various changes and modifications can be made therein by those skilled in the art without departing from the scope and spirit of the invention. All such changes and modifications are intended to be included within the scope of the invention as claimed in the appended claims.

[0177] It should be noted that, in this document, relational terms such as "first" and "second" are used merely 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 a process, method, article, or apparatus. Without further limitations, an element defined by the statement "comprising a…" does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0178] Although the description of the invention has been given in conjunction with the specific embodiments described above, it will be apparent to those skilled in the art that many substitutions, modifications, and variations can be made based on the foregoing. Therefore, all such substitutions, modifications, and variations are included within the spirit and scope of the appended claims.

Claims

1. A method for actively detecting the load-carrying capacity of power supply in a distribution network based on a dynamic load model, characterized in that, include: Set the energy storage unit type for the backup power supply of the distribution network, and retrieve the characteristic quantities of the corresponding energy storage unit type; The initial strategy configuration for detecting the load-carrying capacity of the backup power supply in the distribution network is completed based on the aforementioned characteristic quantities; the characteristic quantities include the charging and discharging characteristics, internal resistance variation law, impact load tolerance threshold, capacity decay characteristics, and constant current output characteristics of each type of energy storage unit. Based on the initial strategy configuration, the ledger information and voltage and current operating parameters of the backup power supply of the distribution network are collected, and a programmable dynamic load model adapted to the energy storage unit type is constructed. The operating parameters such as voltage and current of the backup power supply in the distribution network are collected in real time and input into the programmable dynamic load model to evaluate the load capacity and output a dynamic adjustment strategy for the load capacity. Based on the energy storage unit type, initial strategy configuration and dynamic adjustment strategy of load capacity, the load array is optimized. Active load detection is performed on the optimized backup power supply to generate a final evaluation report of the load capacity of the distribution network backup power supply, which is then visualized on the human-machine interface.

2. The method for active detection of power supply load capacity in distribution networks based on a dynamic load model according to claim 1, characterized in that, The initial strategy configuration for detecting the load-carrying capacity of backup power supplies in the distribution network based on the aforementioned feature quantities includes: Extract the core parameters of the characteristic quantities corresponding to the energy storage unit type to form a structured set of characteristic quantity parameters; The set of feature parameters is integrated into the core detection algorithm. The dynamic load model construction algorithm, load capacity assessment algorithm and load array control algorithm are modified in a targeted manner. The parameter identification logic, load assessment weight and driving threshold boundary of each algorithm are adjusted so that the operation logic of the core detection algorithm is adapted to the inherent operating characteristics of the current energy storage unit. Based on the sensitivity of energy storage unit parameters reflected by the feature parameter set, a differentiated voltage and current synchronous acquisition strategy is configured to determine the A / D sampling accuracy, sampling rate, filtering method, signal transformation rules and amplitude limiting protection threshold, so as to adapt to the signal acquisition requirements of the energy storage unit. Based on the energy storage unit's impact load tolerance threshold and constant current output characteristics, the basic parameters for load loading are preset, and the initial loading amplitude, single loading duration, loading steps, loading interval, and switch opening and closing cyclic loading simulation frequency are determined. The initial loading parameters do not exceed the safety tolerance boundary of the energy storage unit. Based on the capacity decay characteristics and overcurrent and overvoltage tolerance thresholds of the energy storage unit, a multi-level safety protection strategy is set during the detection process, and the load derating ratio, shutdown triggering conditions and emergency handling rules under abnormal operating conditions are clarified. The optimized core detection algorithm, customized data acquisition strategy, preset load loading strategy, and set security protection strategy are integrated into a unified initial strategy set and solidified into the computing processor of the detection device, serving as the basis for subsequent data acquisition, model building, load capacity assessment, and load array control.

3. The method for active detection of power supply load capacity in distribution networks based on a dynamic load model according to claim 2, characterized in that, The step of collecting the ledger information and voltage and current operating parameters of the backup power supply in the distribution network based on the initial strategy configuration, and constructing a programmable dynamic load model adapted to this type of energy storage unit, includes: The basic ledger information and real-time operating parameters of the power distribution network backup power supply are collected according to the customized data acquisition strategy. The ledger information includes the rated voltage, rated current, rated capacity, operation and maintenance records and historical load characteristics of the power distribution network backup power supply. The real-time operating parameters are voltage and current analog signals synchronously collected by high-precision sensors. The algorithm for constructing a dynamic load model optimized by the energy storage unit features in the initial strategy configuration is retrieved, the preset polynomial dynamic load basic model is loaded, a time series data cache array is created, the above-mentioned data sources are imported into the cache array, and data preprocessing is completed by sliding window mean filtering. The load characteristic parameters of the backup power supply in the distribution network under the historical stable operation state are read. Combined with the pre-processed real-time data source, the recursive least squares (RLS) identification algorithm is used to iteratively solve the linear model in the preset polynomial dynamic load basic model. During the iteration process, the number of parameter identification iterations matching the energy storage unit type is set to obtain the optimal load characteristic parameters that are suitable for the current backup power supply of the distribution network under test. Dynamic load characteristics are calculated based on the optimal load characteristic parameters. At the same time, the energy storage unit characteristic quantities are incorporated to make targeted dynamic corrections to the model, so as to obtain the final programmable dynamic load model.

4. The method for active detection of power supply load capacity in distribution networks based on a dynamic load model according to claim 3, characterized in that, Before importing the aforementioned data sources into the cache array, the process also includes: The acquired voltage and current analog signals are sequentially processed by signal transformation, low-pass filtering, and amplitude limiting protection. The backup power supply voltage is transformed into a standard linear voltage of 0~3.3V through linear resistor voltage division, linear opto-isolation, and operational amplification. Then, it is converted into a digital quantity by A / D conversion with the accuracy set in the initial strategy configuration, forming a structured detection data source.

5. The method for active detection of power supply load capacity in distribution networks based on a dynamic load model according to claim 4, characterized in that, The calculation of dynamic load characteristics based on the optimal load characteristic parameters includes: Calculate the change value of dynamic load impedance and load change rate of the backup power supply of the distribution network over time. The dynamic load impedance reflects the impedance change law of the load under normal and transient conditions, and the load change rate reflects the dynamic change speed of the load.

6. The method for active detection of power supply load capacity in distribution networks based on a dynamic load model according to claim 5, characterized in that, The load array optimization is performed by combining the energy storage unit type, initial strategy configuration, and dynamic load capacity adjustment strategy. Active load detection is then performed on the optimized backup power supply, including: Based on the characteristic boundary of the energy storage unit type, the load array control algorithm fixed in the initial strategy configuration, and the loading parameter requirements in the dynamic adjustment strategy of load capacity, the linear load array composed of series and parallel power transistors is dynamically combined and optimized to adjust the number of series and parallel power transistors and the connection method. Based on the drive signal of the dynamic closed-loop control circuit, the optimized linear load array is driven to perform active loading detection on the power supply backup power supply without shutdown. During the detection process, the real-time response parameters such as voltage, current and temperature of the power supply backup power supply and the actual output load parameters of the linear load array are collected simultaneously to identify the load performance and response characteristics of the power supply backup power supply under different load conditions in real time.

7. The method for active detection of power supply load capacity in distribution networks based on a dynamic load model according to claim 5, characterized in that, The active load detection of the optimized backup power supply also includes: comparing the collected response parameters with the multi-level safety protection thresholds set in the initial strategy configuration in real time during the load detection process; if the detected parameters exceed the warning threshold, immediately perform load derating operation according to the dynamic adjustment strategy of load capacity; if the parameters exceed the termination threshold, immediately trigger the load termination command, stop the active load detection and generate abnormal warning information.

8. A distribution network power supply load-carrying capacity active detection circuit based on a load dynamic model, wherein the circuit employs a distribution network power supply load-carrying capacity active detection method based on a load dynamic model as described in any one of claims 1-7, characterized in that... The circuit includes: Data acquisition circuit module: includes signal conversion circuit, low-pass filter circuit and amplitude limiting protection circuit; used for synchronous acquisition, conversion, filtering and protection of voltage and current signals, and outputting digital signals to the arithmetic processing circuit module; The arithmetic processing circuit module includes peripheral basic circuits, A / D conversion circuits, chip main control circuits, and DAC interface driver circuits; the chip main control circuit has built-in data processing programs, dynamic load model algorithm programs, load capacity assessment algorithm programs, and logic control programs. The load drive circuit module includes an amplifier circuit, a differential feedback circuit, and a comparator circuit. The amplifier circuit is composed of an operational amplifier, which amplifies the 0~3.3V low voltage signal input from the operational processing circuit module into a 0~12V drive signal. The differential feedback circuit acquires the real-time current signal of the linear load array circuit module and feeds it back to the inverting input of the comparator circuit. The comparator circuit compares the amplified drive signal with the current feedback signal in real time to form a dynamic closed-loop control circuit and outputs the drive signal to the linear load array circuit module. The linear load array circuit module consists of several low-switching-loss high-speed IGBT power transistors, a manganese-copper shunt, and a current feedback circuit. Each IGBT power transistor is connected in parallel with a fast recovery full-current diode. Multiple IGBT power transistors are connected in series and parallel to form a programmable linear load array. The load current is linearly regulated by the gate signal. The manganese-copper shunt is connected in series with the load array and uses micro-voltage acquisition to achieve real-time load current detection. The current signal is then transmitted to the differential feedback circuit of the load drive circuit module via the current feedback circuit. The human-computer interaction circuit module includes a TTL driver interface circuit and a touch serial port display circuit. The TTL driver interface circuit enables bidirectional data interaction with the arithmetic processing circuit module, receiving detection data, evaluation results, and control commands and transmitting them to the touch serial port display circuit. The touch serial port display circuit is a serial port display circuit with a touch screen, which completes the functions of setting detection parameters, real-time display of detection status, visualization of load capacity evaluation results, and issuing manual commands.

9. The active detection circuit for the load-carrying capacity of power distribution network based on a load dynamic model according to claim 8, characterized in that, Also includes: The AT32F chip in the computing circuit module also integrates a logic judgment program unit and a human-machine interface control program unit. The logic judgment program unit completes the threshold judgment of the running parameters and the logic triggering of abnormal working conditions during the detection process. The human-machine interface control program unit realizes the instruction parsing, status feedback and dynamic interface update of the touch serial port display circuit.

10. A handheld rapid testing tool, characterized in that, The handheld rapid testing tool adopts an active detection circuit for the load-carrying capacity of power distribution networks based on a load dynamic model, as described in any one of claims 8-9.

Images