Intelligent collaborative switching control system and method for multi-diameter valve flow channel

Abstract

The application discloses an intelligent collaborative switching control system and method for a multi-diameter valve flow channel, which collects full working condition operation data of a multi-diameter valve system, extracts valve switching characteristic parameters, and establishes a flow channel mapping relationship between diameters; based on the mapping relationship and the switching characteristics, a cascade action chain containing time sequence dependence is constructed, and the action sequence and trigger conditions of each diameter are determined; by analyzing the pressure distribution and flow velocity change in the cascade action chain, water hammer strength is predicted, reverse pressure wave parameters are designed, and water hammer offset control instructions are generated; the high and low pressure area distribution in the flow field is identified, a flow field complementary collaborative scheme is constructed, and a pressure balance control strategy is generated; the superposition elimination point is determined by analyzing the pressure wave propagation path, the pressure pulse resonance characteristics are extracted, and time sequence control parameters are generated; the time sequence control parameters are matched with the cascade action chain, the synchronous control node is determined by delay compensation, and the time-sharing control sequence is generated to realize intelligent collaborative switching of the multi-diameter valve flow channel.

Description

Technical Field

[0001] This invention relates to the field of fluid control technology, and in particular to an intelligent coordinated switching control system and method for multi-bore valve flow channels. Background Technology

[0002] In complex fluid transport systems, multi-bore valves are widely used in petrochemical, natural gas transportation, and water conservancy projects, undertaking critical functions such as flow distribution, pressure regulation, and flow direction switching. However, traditional valve control methods typically employ independent control strategies, with each valve operating independently based on local measurement signals, lacking an overall coordination mechanism. This control approach is prone to problems such as sudden pressure changes, flow fluctuations, and pipeline vibrations during valve operation, which not only affect system stability but may also cause safety hazards such as equipment damage.

[0003] Existing technologies have several limitations when dealing with multi-valve systems: they cannot effectively predict and control transient flow phenomena caused by valve action; they struggle to coordinate pressure and flow rates between different flow channels; they lack comprehensive analysis and control capabilities for the system's dynamic response characteristics; and the timing of valve action is not optimized, easily leading to unfavorable interactions. Therefore, there is an urgent need to develop a new control method capable of coordinating the actions of multiple valves, effectively controlling transient processes, and optimizing overall system performance, thereby improving the safety and operational efficiency of fluid transport systems. Summary of the Invention

[0004] This invention provides an intelligent collaborative switching control system and method for multi-bore valve flow channels. By establishing a cascade relationship of valve actions and a timing optimization mechanism, collaborative control of multi-bore valves is achieved. This invention collects system operation data and extracts switching features to construct a cascaded action chain with timing dependencies; predicts water hammer intensity through pressure and flow velocity analysis and designs a reverse pressure wave to achieve active cancellation; identifies pressure distribution characteristics to construct a flow field complementary scheme; generates timing control parameters based on resonance characteristics; and completes intelligent collaborative switching control through delay compensation and synchronization node establishment.

[0005] The first aspect of this invention proposes an intelligent coordinated switching control system and method for multi-bore valve flow channels, comprising:

[0006] Real-time fluid parameters of multi-bore valves are collected, and feature analysis is performed on the real-time fluid parameters to identify the current flow channel working mode. The flow channel working mode is then used to construct a flow channel state transition matrix.

[0007] The candidate path set is generated by reverse path deduction through the flow channel state transition matrix. Pressure wave time series analysis is performed on the candidate path set to identify leverage switching points. A time series leverage switching sequence is generated based on the leverage switching points. A cascaded action chain is constructed based on the time series leverage switching sequence.

[0008] The pressure gradient distribution and velocity change curve of each path are extracted from the cascade action chain. The water hammer intensity distribution is predicted based on the pressure gradient distribution. The water hammer occurrence sequence is determined through the velocity change curve. The reverse pressure wave parameters are designed by combining the water hammer intensity distribution and the water hammer occurrence sequence. The water hammer cancellation control command is generated using the reverse pressure wave parameters.

[0009] Obtain the pressure distribution characteristics of each path in the cascade action chain, identify high-pressure and low-pressure areas based on the pressure distribution characteristics, construct a flow field complementary and cooperative scheme based on the distribution of the high-pressure and low-pressure areas, and generate a pressure balance control strategy using the flow field complementary and cooperative scheme.

[0010] Based on the water hammer cancellation control command and the pressure balance control strategy, a multi-valve collaborative control timing sequence is constructed. Pressure wave propagation path analysis is performed through the multi-valve collaborative control timing sequence to determine the pressure wave superposition elimination point. Pressure pulse resonance features are extracted from the pressure wave superposition elimination point. Timing control parameters are generated based on the pressure pulse resonance features.

[0011] The timing control parameters are matched with the cascaded action chain to generate synchronous control nodes. The action time of each flow path is determined based on the synchronous control nodes. A time-sharing control sequence is generated according to the action time to complete the intelligent coordinated switching control of the flow path of the multi-flow path valve.

[0012] A second aspect of this invention proposes an intelligent coordinated switching control system for multi-bore valve flow channels, comprising:

[0013] The flow state identification module is used to collect real-time fluid parameters of multi-bore valves, perform feature analysis on the real-time fluid parameters to identify the current flow channel working mode, and construct a flow channel state transition matrix using the flow channel working mode;

[0014] The path planning module is used to generate a candidate path set by performing reverse path deduction through the flow channel state transition matrix, perform pressure wave time series analysis on the candidate path set to identify leverage switching points, generate a time series leverage switching sequence based on the leverage switching points, and construct a cascaded action chain based on the time series leverage switching sequence.

[0015] The water hammer control module is used to extract the pressure gradient distribution and flow velocity change curve of each diameter from the cascade action chain, predict the water hammer intensity distribution based on the pressure gradient distribution, determine the water hammer occurrence sequence through the flow velocity change curve, design reverse pressure wave parameters by combining the water hammer intensity distribution and the water hammer occurrence sequence, and generate water hammer cancellation control commands using the reverse pressure wave parameters.

[0016] The collaborative control module is used to acquire the pressure distribution characteristics of each path in the cascade action chain, identify high-pressure and low-pressure areas based on the pressure distribution characteristics, construct a flow field complementary collaborative scheme based on the distribution of the high-pressure and low-pressure areas, and generate a pressure balance control strategy using the flow field complementary collaborative scheme.

[0017] The timing optimization module is used to construct a multi-valve collaborative control timing based on the water hammer cancellation control command and the pressure balance control strategy, analyze the pressure wave propagation path through the multi-valve collaborative control timing to determine the pressure wave superposition elimination point, extract pressure pulse resonance features from the pressure wave superposition elimination point, and generate timing control parameters based on the pressure pulse resonance features.

[0018] The execution control module is used to perform timing matching between the timing control parameters and the cascaded action chain to generate synchronous control nodes, determine the action time of each flow path based on the synchronous control nodes, generate a time-sharing control sequence according to the action time, and complete the intelligent coordinated switching control of the flow path of the multi-flow-path valve.

[0019] The beneficial effects of this invention are reflected in the following points: First, by constructing a cascaded action chain, the timing management of multi-bore valve actions is realized. The action sequence and triggering conditions of each valve are determined based on the flow channel mapping relationship and switching characteristics, transforming the originally independent control actions into an ordered cascaded process, reducing mutual interference and pressure disturbances during valve switching. Second, an active control strategy is adopted to cope with transient pressure changes. By predicting the water hammer intensity and generating reverse pressure wave parameters, destructive interference is generated before the pressure impact arrives. Simultaneously, a complementary and coordinated flow field scheme is constructed based on the distribution of high and low pressure zones, utilizing the pressure potential energy difference to achieve natural flow and reduce system energy consumption. Finally, by extracting pressure pulse resonance characteristics to optimize the control timing, the system's natural frequency is avoided. Synchronous control nodes are established and delay compensation is performed to ensure precise coordination of multi-valve actions. The generated time-sharing control sequence can be applied to industrial control systems, improving the automation level and operational safety of multi-bore valve systems.

[0020] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0021] The accompanying drawings illustrate specific examples of the technical solutions described in this invention and, together with the detailed embodiments, form part of the specification, serving to explain the technical solutions, principles, and effects of this invention.

[0022] Unless otherwise specified or defined, the same reference numerals in different figures represent the same or similar technical features, and different reference numerals may be used to represent the same or similar technical features.

[0023] Figure 1 This is a flowchart illustrating an intelligent collaborative switching control system and method for multi-bore valve flow channels according to the present invention.

[0024] Figure 2 This is a structural block diagram of an intelligent collaborative switching control system for multi-bore valve flow channels according to the present invention. Detailed Implementation

[0025] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0026] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0027] The technical solutions of the embodiments of this application will be described below.

[0028] like Figure 1 As shown, this embodiment of the invention provides an intelligent collaborative switching control system and method for multi-bore valve flow channels, including the following steps S110 to S160:

[0029] Step S110: Collect real-time fluid parameters of the multi-bore valve, perform feature analysis on the real-time fluid parameters to identify the current flow channel working mode, and construct the flow channel state transition matrix using the flow channel working mode.

[0030] Specifically, real-time fluid parameters are collected by sensor arrays installed at the inlet, outlet, and key nodes of each flow channel in a multi-bore valve. Fluid parameter acquisition includes basic physical quantities such as pressure, flow rate, velocity, and temperature. Pressure sensors are positioned at the inlet and outlet of each flow channel, with a sampling frequency set to 100Hz to capture pressure fluctuation details. Flow meters are installed on the centerline of the main flow channels, achieving a measurement accuracy of ±0.5%. Temperature sensors are used to monitor the impact of fluid temperature changes on flow characteristics. The data acquisition system adopts a distributed architecture, with each sensor connected to the data acquisition unit via a fieldbus for multi-point synchronous sampling. The acquired raw data undergoes filtering to eliminate high-frequency noise, and a moving average method is used to smooth the data curves, with a window length set to 10 sampling points. Data preprocessing also includes outlier removal and missing value imputation to ensure data quality. A time synchronization mechanism ensures the temporal consistency of data at each measurement point, with timestamp accuracy reaching the millisecond level. The real-time data cache uses a circular buffer structure, retaining historical data from the most recent 5 minutes. Multi-point collaborative acquisition yielded complete real-time fluid parameters, including pressure time series P(t), flow time series Q(t), velocity distribution v(x,t), and temperature field T(x,t).

[0031] Multi-dimensional feature analysis is performed on the collected real-time fluid parameters to identify the current flow channel operating mode. Feature extraction is performed by calculating the average pressure, pressure fluctuation amplitude, pressure gradient, and pulsation frequency from the pressure time series P(t); extracting the instantaneous flow rate, average flow rate, and flow distribution ratio r_i = Q_i / Q_total from the flow rate time series Q(t), where Q_i is the flow rate of the i-th channel and Q_total is the total flow rate; evaluating the flow velocity uniformity index from the velocity distribution v(x,t); and analyzing the influence of the temperature gradient on the flow from the temperature field T(x,t). Time-domain analysis identifies periodic patterns within a 5-minute historical window, and frequency-domain analysis extracts the main frequency components through fast Fourier transform. The extracted multi-dimensional feature parameters, such as pressure, flow rate, velocity, and temperature, are reduced in dimensionality through principal component analysis. The top four principal components with a contribution rate exceeding 85% are selected to construct the feature vector F = [f_1, f_2, f_3, f_4], where f_i represents the i-th principal component. Based on the feature vector F, a pre-trained support vector machine classifier is used to identify the operating mode. The operating modes are divided into single-path mode (M1), dual-path parallel mode (M2), multi-path splitting mode (M3), and switching transition mode (M4). Sub-modes are further subdivided into operating conditions; for example, M2 is divided into uniform splitting (M2a) and master-slave splitting (M2b). The real-time identification process runs continuously, updating every 100ms to obtain the current flow channel operating mode M(t).

[0032] A flow channel state transition matrix is ​​constructed using a sequence of continuously identified flow channel operating modes to quantify the transition relationships between different operating modes. The state set is defined as S = {M1, M2a, M2b, M3, M4}, containing all major operating modes and key sub-modes. A transition event sequence is constructed by recording mode changes at consecutive time points. When mode M(t_k) at time t_k transforms into mode M(t_k+1) at time t_{k+1}, a transition from state i to state j is recorded. The number of occurrences of each type of transition event N_{ij} within the time window is counted, where i represents the initial state and j represents the target state. The transition probability is calculated as p_{ij} = N_{ij} / Σ_j N_{ij}, which is the number of transitions from state i to state j divided by the total number of transitions starting from state i. A 5×5 state transition matrix P is constructed, where the matrix element p_{ij} represents the probability of transitioning from state i to state j. Hazardous transfer paths are identified using expert knowledge. For example, a direct jump from M1 to M3 might cause severe water hammer, so the corresponding p_{13} is set to 0. Matrix normalization ensures that the sum of elements in each row is 1, i.e., Σ_j p_{ij} = 1, satisfying probability distribution constraints. Time-varying characteristics are implemented using a sliding window mechanism, updating matrix parameters hourly to adapt to changes in operating conditions. Steady-state analysis is performed by solving π = πP to obtain the steady-state probability vector π, where π_i represents the probability of the system being in state i during long-term operation. After statistical modeling and probability normalization, the flow channel state transition matrix P is constructed.

[0033] Step S120: Generate a candidate path set by reverse path deduction through the flow channel state transition matrix, identify the leverage switching points by pressure wave time series analysis of the candidate path set, generate a time series leverage switching sequence based on the leverage switching points, and construct a cascade action chain based on the time series leverage switching sequence.

[0034] In some embodiments, the step of generating a candidate path set by reverse path deduction through the flow channel state transition matrix includes: extracting the switching cost between states based on the flow channel state transition matrix; obtaining the total path cost by inversely accumulating the switching cost from the target state based on the switching cost; and selecting multiple paths with the minimum cost based on the total path cost to form a candidate path set.

[0035] Based on the flow path state transition matrix P, the switching costs between any two state pairs are systematically extracted. The cost calculation comprehensively considers multiple influencing factors during the switching process. The pressure shock cost is obtained by estimating the maximum pressure surge caused by the state transition, using the water hammer formula ΔP = ρ×c×Δv to predict the shock intensity. The flow disturbance cost quantifies the instability of the flow rate during the switching process, obtained by calculating the integral of the flow rate change rate. The energy loss cost assesses the irreversible energy dissipation during the switching process, including valve throttling losses and pipe friction losses. The time cost reflects the stabilization time required to complete the state transition; rapid and smooth switching has a lower time cost. The comprehensive cost is obtained through weighted summation, with weighting coefficients set according to the priority of specific operating conditions. Special switching paths, such as emergency shutdowns, require additional safety cost assessments. State pairs with a probability of zero in the transition matrix P are assigned an infinite cost, indicating that switching is prohibited. Standardization of cost values ​​ensures the comparability of different types of costs.

[0036] Based on the extracted switching costs, backward dynamic programming is performed to calculate the total path cost, starting from the set target state. The algorithm initializes the target state with an arrival cost of zero, indicating that the target has been reached without additional cost. The first round of calculation processes all predecessor states that can directly reach the target state, updating their arrival costs to the cost of a direct switch. The iterative process propagates forward; for any state, its optimal arrival cost is determined by examining all reachable successor states, selecting the path with the minimum cost. The optimal decision records the best successor state selected for each state, used for path reconstruction. Convergence is determined by comparing the cost changes of adjacent iterations; iteration stops when the cost update of all states is less than a threshold. Multiple starting points are handled by calculating the arrival cost of each possible starting state separately. Circular dependencies are broken by introducing virtual nodes to ensure algorithm convergence. Special handling of boundary states ensures computational integrity. After backward accumulation, the minimum total cost distribution from each starting state to the target state is obtained.

[0037] Based on the calculated total path cost, a multi-criteria screening process is performed to generate a high-quality candidate path set. The primary screening criterion is minimizing the total cost; all possible paths are sorted in ascending order of J_total, and the top 10 paths with the lowest costs are initially selected. Path reconstruction is performed by tracing the optimal decision chain from the starting state to the target state, recording the complete state transition sequence. Path length constraints eliminate redundant paths exceeding 8 steps to avoid excessive execution complexity. Path diversity assessment is achieved by calculating the edit distance between paths, ensuring that the candidate set includes different types of switching strategies. Key state coverage checks prioritize paths that pass through important intermediate modes. Time feasibility verification ensures that the cumulative switching time per step does not exceed the total time limit requirement. Risk assessment identifies paths containing high-risk switching and appropriately distributes them. Stability scoring considers the smoothness of each switching step in the path. Finally, the top 5 paths with the highest comprehensive scores are selected as the candidate set. Each candidate path fully records key information such as the state sequence, step cost, and cumulative time, forming the candidate path set.

[0038] In some embodiments, the step of performing pressure wave time-series analysis on the candidate path set to identify leverage switching points includes: extracting the pressure wave propagation characteristics of each path based on the candidate path set; performing time-series energy analysis on the pressure wave propagation characteristics to obtain energy overlap regions; and determining leverage switching points through the energy overlap regions.

[0039] Based on the state switching sequence of each candidate path, an abstract state is converted into specific valve opening and closing actions through a pre-established state-valve mapping table, thereby extracting the propagation characteristic parameters of the pressure wave. The state mapping table defines the valve operation sequence corresponding to each mode switch; for example, switching from M1 to M2a requires opening valve V2 and adjusting valve V3. Pressure wave generation mechanism analysis determines the initial disturbance intensity caused by each state switch, and the pressure jump value is calculated using fluid dynamics equations. Propagation path mapping transforms the abstract state switch into specific valve actions and pipeline flow changes. The mapping process is achieved by querying a pre-established state-pipeline topology association database, which records the changes in pipeline connectivity and corresponding flow path changes caused by each state switch. Wave velocity calculation considers the differences in material, diameter, and fluid characteristics of different pipe sections, forming a segmented wave velocity distribution map. The propagation time matrix records the propagation delay from each disturbance source to each monitoring point. Reflection point identification is completed through pipeline topology analysis, including valve locations, pipe diameter abrupt changes, and tee joints. Reflection intensity is calculated based on the acoustic impedance mismatch, forming a reflection coefficient distribution. Attenuation characteristics are obtained by fitting historical data to obtain an empirical attenuation curve. Phase relationship analysis is used to determine the relative phases of pressure waves generated from different sources; in-phase superposition enhances pressure waves, while out-of-phase superposition weakens them. Spectral characteristics are used to extract the dominant frequency components of the pressure waves. A complete set of propagation characteristic parameters covers the entire process of pressure wave generation, propagation, reflection, and attenuation.

[0040] Temporal energy analysis was performed on the extracted pressure wave propagation characteristics to identify the spatiotemporal energy distribution and interactions of different pressure waves. Energy density calculation converted pressure fluctuations into energy per unit volume. Temporal unfolding plotted the pressure wave energy generated by each switching action on the time axis, forming an energy evolution curve. Spatial distribution analysis discretized the pipe network using a grid and calculated the energy flux of each grid cell. Multi-source superposition used the linear superposition principle to calculate the total energy field. Identification of energy overlap regions was achieved by setting an energy density threshold; regions exceeding the threshold were marked as high-energy areas. Time window analysis searched for the overlap of energy peaks within a range of ±100 ms, with the overlap degree quantified using a cross-correlation function. Positive overlap produced an energy enhancement effect, with energy reaching 1.6-2.0 times that of a single wave. Negative overlap led to energy cancellation, forming energy troughs. The stability of the overlap region was evaluated by its duration; a stable overlap region lasting longer than 50 ms was considered. The spatial extent was determined by the energy diffusion radius, typically 2-3 times the pipe diameter. The spatiotemporal coordinate set of the energy overlap region constituted a candidate region for leverage.

[0041] By analyzing the identified energy overlap areas, specific leveraged switching points are determined. The leverage criterion is set as the overlap energy exceeding 1.5 times the single-wave energy and lasting for more than 30ms. This threshold considers a safety margin of 1.6-2.0 times the theoretical maximum enhancement effect. The spatial location of the switching point is selected near the node with the highest energy density, typically corresponding to critical valves or diversion points. The timing is positioned in the mid-to-late stage of the energy rise edge, when energy accumulation is sufficient but has not yet begun to decay. Leverage strength is rated into three levels: strong, medium, and weak, based on the available energy. Directional matching checks ensure that the leverage direction matches the desired switching direction, avoiding reverse obstruction. Stability scoring considers the temporal stability and spatial concentration of energy overlap. Accessibility verification checks the physical feasibility of performing the switching at this spatiotemporal point. Prioritization ranking comprehensively considers leverage strength, stability, and accessibility to generate a ranking list. Two to four high-quality leveraged switching points are identified for each path, and their location, time, energy, and level information are recorded. The identification of leveraged switching points determines the key control points for energy utilization.

[0042] Based on the identified available switching points, an optimized timing-based switching sequence is generated. The primary principle of sequence generation is to maximize energy utilization efficiency, decomposing the original continuous switching into discrete actions executed at available switching points. Timing alignment is achieved by adjusting the switching times to match the available switching points, allowing for a time window flexibility of ±20ms. Action decomposition refines complex state transitions into multiple sub-actions, matching and allocating energy based on the energy requirements of each sub-action and the energy supply of available switching points. A sub-action can utilize one or more nearby available switching points. The energy allocation strategy matches the energy requirements of each sub-action with the supply capacity of available switching points, ensuring a balance between supply and demand. Switching sequence optimization considers the mutual influence of preceding and following actions, ensuring that the remaining energy of preceding actions can be utilized by subsequent actions. Time interval settings guarantee sufficient stabilization time between adjacent actions, typically 100-200ms. Sequence integrity checks ensure that all necessary state transitions are covered. Redundant action identification eliminates unnecessary intermediate switching, simplifying the execution sequence. Alternative solutions are generated for each critical action in the main sequence, preparing alternative available switching points. The final timing-based energy transfer sequence includes information on action, time, location, and borrowed energy.

[0043] In some embodiments, constructing a cascaded action chain based on the time-series leveraged switching sequence includes: extracting the remaining energy value of each switching point based on the time-series leveraged switching sequence; identifying an available energy window based on the remaining energy value; determining the optimal triggering time based on the energy window; and connecting the time-series leveraged switching sequence in series according to the optimal triggering time to form a cascaded action chain.

[0044] Based on the energy utilization at each switching point in the time-series switching sequence, the remaining energy distribution after execution is calculated. The remaining energy value E_residual = E_assist - E_consumed is obtained by subtracting the actual consumption from the available energy. The consumed energy is estimated based on the resistance characteristics of the switching action. Energy transfer efficiency considers the transfer loss of the remaining energy from the previous action downstream, with a typical efficiency of 0.7-0.85. The spatiotemporal diffusion model predicts the propagation range and attenuation process of the remaining energy. The quality assessment of the remaining energy distinguishes between high-quality and low-quality energy. The cumulative effect analysis analyzes the superposition of the remaining energy from multiple actions to form a usable energy pool. Energy lifetime assessment evaluates the effective utilization time window of the remaining energy; energy exceeding the time limit is no longer usable. The spatial distribution map shows the distribution density of the remaining energy in the pipeline network. Directional analysis determines the main flow direction of the remaining energy. Energy account management tracks the energy balance of each node. The remaining energy analysis forms a complete spatiotemporal energy distribution map E_residual(x,t). Based on the income and expenditure records of energy account management, a continuous spatiotemporal energy distribution function E_residual(x,t) is generated through spatial interpolation and temporal smoothing algorithms. The spatial dimension uses bilinear interpolation of the energy values ​​of grid nodes, and the temporal dimension uses cubic spline interpolation to ensure the continuity of the distribution function.

[0045] Based on the spatiotemporal distribution of remaining energy E_residual(x,t), usable energy windows are identified. An energy window is defined as a spatiotemporal region where the remaining energy density exceeds a utilization threshold. Temporally, the window begins when energy accumulation exceeds the threshold and ends when energy decay falls below the threshold. Spatially, the window covers a network area with sufficient energy density, typically a limited area centered on the energy source. Window effectiveness is evaluated considering energy density, duration, and spatial coverage. Overlapping windows are merged using a union operation to form a larger usable area. Window stability is assessed using an energy fluctuation coefficient; fluctuations less than 20% are considered stable. Accessibility constraints eliminate physically unusable window regions. Window capacity estimates determine the maximum energy requirement for supported actions. Timing constraints ensure that the window's occurrence time matches the planned action time. Priority ranking determines the utilization order based on window quality and matching degree.

[0046] Based on the identified energy window set, an optimization algorithm determines the optimal triggering time for each action. The goal of triggering time selection is to maximize energy utilization and action success rate. Energy utilization is measured by the ratio of actual acquired energy to the available energy in the window. Action success rate considers the reliability and stability of executing the action at that time. Time constraints ensure that the triggering time is within the validity period of the energy window W_i, with a 20-30ms safety margin to prevent boundary effects. Multi-action coordination is achieved by establishing a temporal dependency graph to ensure reasonable intervals between actions. Trigger point selection tends to be on the stable segment of the energy rising edge, avoiding instability at peak points. Spatial matching ensures the consistency between the triggering position and the action execution position. Conflict detection prevents multiple actions from competing for the same energy window. A dynamic adjustment mechanism allows for fine-tuning of the triggering time based on real-time status. Robustness analysis evaluates the sensitivity of the triggering time to disturbances. The optimal triggering time set {t_trigger_1, t_trigger_2, ..., t_trigger_n} for each action is obtained through optimization calculation.

[0047] Discrete switching actions are chained together according to the determined optimal triggering time {t_trigger_i} to form a cascaded action chain with causal relationships. The cascade structure design makes the output of the preceding action the input of the downstream action, forming a relay transmission of energy and information. The action node N_i is defined with four elements: triggering time t_trigger_i, execution position x_i, energy requirement E_need_i, and output characteristic E_out_i. The connection relationship is represented by two types of edges: energy flow and temporal dependency, forming a directed graph structure. The cascade effect G_ij quantifies the gain effect of node i on node j. Critical path identification uses graph algorithms to find the main path affecting overall performance. Parallel branch processing allows some actions to be executed concurrently, improving efficiency. Synchronization point setting ensures that parallel branches converge at critical nodes. Fault tolerance mechanism inserts checkpoints in the cascade chain, allowing recovery from checkpoints in case of failure. Through the designed cascade structure, efficient energy utilization and coordinated action are achieved, ultimately constructing a cascaded action chain C = {N_1→N_2→...→N_n}.

[0048] Step S130: Extract the pressure gradient distribution and velocity change curve of each path from the cascade action chain, predict the water hammer intensity distribution based on the pressure gradient distribution, determine the water hammer occurrence sequence through the velocity change curve, design the reverse pressure wave parameters by combining the water hammer intensity distribution and the water hammer occurrence sequence, and generate water hammer cancellation control commands using the reverse pressure wave parameters.

[0049] Specifically, based on the constructed cascaded action chain, the changes in flow parameters generated in each pipe during the execution of each action node are analyzed. The switching information contained in the action node directly affects the flow field distribution, including the trigger time, execution location, and energy magnitude. Pressure gradient extraction is performed through virtual monitoring points arranged in each pipe. The pressure values ​​of the virtual monitoring points are calculated using a fluid dynamics interpolation algorithm based on the actual sensor data collected in step S110. The spacing between monitoring points is set to 5-10 times the pipe diameter. The pressure changes caused by the action of each node propagate in the pipe, forming a dynamic pressure field. The pressure gradient calculation ∇P =∂P / ∂x adopts the central difference scheme to improve the calculation accuracy. The velocity change curve is correlated with the pressure change through the Bernoulli equation, Δv = √(2×ΔP / ρ). The coupling effect of multiple pipes is described by the mass conservation and momentum conservation equations. The time resolution is set to 10 ms to capture fast transient processes. The spatial interpolation uses the cubic spline method to ensure the continuity of the gradient. The cumulative effect of cascading actions is calculated by superimposing the contributions of each node. The propagation delay is handled using a time-weighted superposition method, which determines the delay time based on the triggering time and propagation distance of each action node to ensure the timing accuracy of the pressure gradient calculation. Based on the extraction process, pressure gradient distribution and velocity variation curves for each path are generated.

[0050] Based on the extracted pressure gradient distribution, water hammer theory is used to predict the potential water hammer intensity for each pipe diameter. Water hammer generation mechanism analysis identifies locations of abrupt pressure gradient changes as potential water hammer sources; areas where the absolute value of the gradient exceeds a critical threshold are marked as hazardous zones. Water hammer intensity is calculated using the Jukowski formula ΔH = a×Δv / g, where ΔH is the water hammer pressure head, a is the pressure wave velocity, Δv is the velocity abrupt change, and g is the gravitational acceleration. The pressure wave velocity considers the combined effects of fluid compressibility and pipe elasticity. The spatial characteristics of the intensity distribution show a decay pattern from the water hammer source towards both ends. The temporal evolution follows a wave equation, with a rapid initial rise followed by gradual decay. The superposition effect of multi-source water hammer is evaluated using the linear superposition principle: H_total = Σ H_i, where H_i is the intensity generated by the i-th water hammer source. Hazard level classification categorizes water hammer intensity into three levels: slight, moderate, and severe, corresponding to different multiples of the working pressure. Pipeline pressure-bearing capacity verification ensures that the predicted maximum pressure does not exceed the allowable value. The water hammer duration is estimated as t_duration = 2L / a, where L is the characteristic length of the pipe and a is the pressure wave velocity.

[0051] By analyzing the temporal characteristics of the flow velocity change curves, the timing of water hammer occurrences in each pipe diameter is determined. Water hammer trigger point identification is achieved by detecting abrupt changes in flow velocity; when the rate of change in flow velocity exceeds a set threshold, it is recorded as the water hammer initiation time t_hammer. The threshold setting considers the normal operating flow velocity change range, typically taking 3-5 times the normal rate of change. The temporal relationship between multiple pipe diameters is determined through cross-correlation analysis; the time difference corresponding to the correlation peaks reflects the water hammer propagation delay. Propagation path tracing starts from the trigger point and calculates the arrival time at each node along the pipe network topology. Wavefront arrival time t_arrival = t_hammer + L / a, where t_hammer is the water hammer initiation time, L is the propagation distance, and a is the pressure wave velocity. The timing of reflected waves is determined by the round-trip time 2L / a. Periodic oscillation identification is achieved through spectral analysis; the dominant frequency f = a / (4L) corresponds to the basic oscillation mode of the pipe, where f is the oscillation frequency. Key moments are marked, including the initial impact, first reflection, and second reflection.

[0052] In some embodiments, the step of designing reverse pressure wave parameters by combining the water hammer intensity distribution and the water hammer occurrence time sequence includes: extracting the peak pressure amplitude based on the water hammer intensity distribution; determining the pressure wave arrival time according to the water hammer occurrence time sequence; calculating the required reverse pressure value based on the peak pressure amplitude; and matching the required reverse pressure value with the pressure wave arrival time to generate reverse pressure wave parameters.

[0053] Based on the predicted water hammer intensity distribution, peak pressure amplitudes at various locations within each pipe diameter are extracted. A peak search algorithm traverses the pressure curve within a time window to identify local maxima. Peak determination criteria include the mathematical condition that the first derivative is zero and the second derivative is negative. For multi-peak cases, the global maximum value is selected as the primary peak, while secondary peaks are recorded for reference. Spatially distributed peak extraction proceeds along the pipe axis, forming a continuous peak envelope. Statistical characteristics of peak values ​​include parameters such as maximum value, average value, and standard deviation, used to comprehensively assess the overall level and distribution characteristics of pressure shocks. Abnormal peaks are identified using the 3σ criterion and specially marked to prevent extreme values ​​from affecting the overall analysis results. The time difference between the peak occurrence and the water hammer triggering time reflects the pressure establishment process from the initial disturbance to reaching the peak. Peak duration assesses the length of time the pressure remains high, directly affecting the design of control strategies. Peak comparisons between different pipe diameters identify the most dangerous locations and areas requiring the most focused protection in the system. A safety margin of 10%-20% is considered on top of the peak value as a reserve.

[0054] Based on the determined water hammer occurrence sequence, the propagation process and arrival time of the pressure wave in the pipeline network are calculated. The propagation model is based on the characteristic line method, tracing the trajectory of the wavefront along the propagation path of the pressure wave. The arrival time calculation needs to consider two different cases: direct wave and reflected wave. The arrival time of the direct wave is t_arrive_direct = t_hammer + L / a, and the arrival time of the reflected wave is t_arrive_reflect = t_hammer + 2L / a, where t_hammer is the water hammer occurrence time, L is the propagation distance, and a is the pressure wave velocity. The complexity of the pipeline network topology is handled using graph theory algorithms to calculate the shortest propagation path between any two nodes. Wave distribution at branch nodes follows the acoustic propagation principle, and the transmission coefficient and reflection coefficient are determined by the impedance matching conditions of the pipeline. Multipath propagation processing requires identifying all possible propagation paths and calculating the arrival time of each path separately. The time accuracy is controlled at the millisecond level, fully meeting the response requirements of the industrial control system. Phase relationship analysis is used to determine the relative phase difference between pressure waves along different paths. The attenuation correction during propagation considers the impact of energy loss on the arrival time. The effects of boundary conditions are adjusted accordingly using the endpoint reflection coefficient.

[0055] Based on the extracted peak pressure amplitude, the required reverse pressure value to generate destructive interference is calculated. The core principle of reverse pressure design is based on the destructive interference phenomenon in physics, where two equal-amplitude, opposite-phase pressure waves can achieve a complete destructive effect when superimposed. The ideal reverse pressure calculation formula is P_counter = -H_peak×α, where P_counter is the reverse pressure value, H_peak is the peak pressure, and α is the compensation coefficient used to account for the non-ideal characteristics of the actual control system. The compensation coefficient is set between 0.8 and 1.0 to ensure sufficient destructive effect while avoiding overcompensation that could generate new reverse water hammer. The segmented compensation strategy decomposes the complete pressure curve into multiple independent segments, each with a corresponding reverse pressure value designed according to its characteristics. Nonlinear correction is used to handle the nonlinear propagation effect of large-amplitude pressure waves. The nonlinear correction formula is P_counter = -H_peak×α×(1+β×H_peak / P_rated), where β is the nonlinear enhancement coefficient to ensure effective cancellation under high pressure, and P_rated is the pipeline rated pressure. Safety constraints ensure that the calculated reverse pressure does not exceed the actual regulating capacity range of the system. Multi-point collaborative design schemes generate reverse pressure simultaneously at multiple key locations in the pipeline network, improving the overall destructive effect. Pressure gradient matching requires not only matching the pressure amplitude but also considering the matching of the pressure change rate.

[0056] The calculated reverse pressure value is precisely spatiotemporally matched with the determined arrival time to generate complete reverse pressure wave parameters. The parameter matching process strictly adheres to the causal principle, ensuring that the reverse pressure is generated at the appropriate time before the water hammer pressure wave reaches the target location. The time advance is set to 50-100ms to compensate for the mechanical response delay of the actuator and the transmission delay of the control signal. Establishing the spatial correspondence precisely aligns the applied position of the reverse pressure with the location where the water hammer peak occurs. The parameter assembly process requires defining four key elements: the spatial location of the control point (e.g., valve number or pipeline mileage), the trigger time, the required pressure value, and the pressure maintenance time. Waveform design requires selecting an appropriate pressure release curve type, commonly including step, ramp, or sine waveforms. Waveform realization generates the required pressure pulse by rapidly adjusting the valve opening; the relationship between valve opening and pressure change is determined by a pre-calibrated characteristic curve. Complex waveforms are achieved by combining multiple step signals, with the duration of each segment calculated based on the frequency characteristics of the desired waveform. The precision of phase control ensures a 180° phase difference between the reverse pressure wave and the forward water hammer wave at the target location. A redundant design incorporates backup parameter sets at critical system locations to enhance control reliability. Each control point is configured with complete action parameters, including when to act, where to act, the magnitude of the action, and the duration; these parameters collectively constitute the water hammer countermeasure execution scheme.

[0057] Using the designed reverse pressure wave parameters, water hammer compensation control commands are generated that can be directly executed by the field control system. The command conversion process maps abstract physical parameters into specific signal formats that the controller can recognize and execute. The core content of the valve control commands includes the opening adjustment amount, adjustment rate, and precise action timing arrangement. The opening is calculated using the functional relationship between the flow coefficient Cv and pressure, with the formula ΔCv = ΔQ / √(ΔP / SG), ensuring the accuracy of flow regulation. The adjustment rate setting fully considers the physical limitations of the actuator. For emergency responses to water hammer compensation, a rapid adjustment mode is adopted, with an opening change rate of 50%-100% per second, ensuring effective pressure regulation within 100ms. Multi-valve collaborative control is achieved through carefully designed timing arrangement and a unified clock synchronization mechanism, using high-precision timestamps and preset trigger delay compensation to ensure that the actions of each valve maintain strict synchronization at the millisecond level. Priority management sets the water hammer compensation command as the highest priority level of the system, ensuring rapid response in emergencies. The design of safety interlock logic prevents dangerous situations such as command conflicts or over-limit operations. The command format strictly adheres to the standard communication protocol of industrial control systems. An execution feedback channel is configured for real-time monitoring of the control command execution effect. A fault handling mechanism ensures that a backup control scheme can be activated promptly in the event of a main command execution failure. The generated water hammer countermeasure control command precisely adjusts valve opening and timing to produce a reverse pressure wave before the water hammer pressure wave arrives, achieving a destructive interference effect, effectively reducing pressure shocks in the pipeline system and protecting equipment safety.

[0058] Step S140: Obtain the pressure distribution characteristics of each path in the cascade action chain, identify the high-pressure zone and the low-pressure zone based on the pressure distribution characteristics, construct a flow field complementary and cooperative scheme based on the distribution of the high-pressure zone and the low-pressure zone, and generate a pressure balance control strategy using the flow field complementary and cooperative scheme.

[0059] Based on the execution of water hammer countermeasures control commands, abstract action nodes in the cascaded action chain are transformed into specific valve opening control sequences. The action transformation process is achieved by querying a pre-established action-valve mapping database, which records the specific valve operating parameters corresponding to each action type. After executing the valve control sequence, a pressure sensor network deployed in step S110 is used to monitor the pressure changes in each pipe in real time, maintaining a sampling frequency of 100Hz to obtain the real-time pressure distribution state of each pipe after the action execution. Pressure distribution feature extraction is completed through the pressure sensor network deployed at key nodes in each pipe. The sensor density is determined according to the complexity of the flow field, with a typical spacing of 5-10 times the pipe diameter. The pressure distribution of each pipe reflects the residual pressure field after water hammer countermeasures control, including steady-state and dynamic components. Feature parameters include pressure mean, pressure gradient, pressure fluctuation amplitude, and pressure distribution skewness. Spatial distribution uses a three-dimensional interpolation method to extend discrete measurement point data into a continuous pressure field. Temporal features are analyzed using a moving-time window with a window width of 500ms and an update frequency of 100ms. The topological structure of the pressure field is identified through isobaric analysis; dense isobaric regions indicate large pressure gradients. Pressure coupling between multiple paths is evaluated using cross-correlation coefficients; a correlation coefficient greater than 0.7 indicates strong coupling. Abnormal pressure points are identified using a local outlier detection algorithm to prevent sensor malfunctions from affecting the analysis. The non-uniformity index of pressure distribution, NI = σ_p / P_avg, quantifies the degree of non-uniformity in pressure distribution, where σ_p is the pressure standard deviation and P_avg is the average pressure. The pressure distribution characteristic data of each path comprehensively describe the pressure state of the flow field after cascading operations.

[0060] Based on the obtained pressure distribution characteristics, high-pressure and low-pressure regions in the flow field are identified through threshold segmentation and cluster analysis. A high-pressure region is defined as a continuous area where the pressure value exceeds the average pressure plus 1.5-2.0 times the standard deviation. A low-pressure region is defined as a continuous area where the pressure value is below the average pressure minus 1.0-1.5 times the standard deviation. Region boundaries are determined using a pressure gradient operator; when the absolute value of the gradient exceeds a set threshold, it is marked as a boundary point. Connectivity analysis merges scattered high-pressure or low-pressure points into complete regions, with the minimum region area set to twice the square of the pipe diameter. Region feature extraction includes center location, area size, average pressure, and pressure extremes. The formation causes of high-pressure regions are identified as flow channel contraction, valve throttling, or multi-flow confluence. The generation mechanisms of low-pressure regions include flow channel expansion, flow splitting points, or vortex regions. Region stability is evaluated through time series analysis; stable regions show a positional change of less than 10% during the observation period. The migration velocity and direction of dynamic regions are obtained through centroid tracking. The mutual influence between regions is evaluated through potential energy analysis; high-pressure regions have a driving effect on low-pressure regions. The high-pressure and low-pressure distribution maps generated by the identification results clearly show the pressure imbalance in the flow field, which provides an opportunity for energy redistribution.

[0061] In some embodiments, constructing a complementary and coordinated flow field scheme based on the distribution of the high-pressure zone and the low-pressure zone includes: extracting the pressure difference value between each region based on the distribution of the high-pressure zone and the low-pressure zone; determining the natural direction of pressure flow based on the pressure difference value; designing auxiliary channels based on the natural direction of pressure flow to obtain an optimized flow channel layout; and forming a complementary and coordinated flow field scheme through the optimized flow channel layout.

[0062] Based on the identified high-pressure and low-pressure zones, the pressure difference between each pair of zones is systematically calculated. The pressure difference calculation formula ΔP_ij = P_h_i - P_l_j - ΔP_loss considers the average pressure P_h_i in the high-pressure zone, the average pressure P_l_j in the low-pressure zone, and the estimated friction loss ΔP_loss between the two zones. The friction loss is estimated using the Darcy-Wiesbach formula ΔP_loss = f × (L / D) × (ρv² / 2), where f is the friction coefficient, L is the flow length, D is the equivalent diameter, ρ is the fluid density, and v is the flow velocity. Effective pressure difference screening eliminates pressure difference pairs less than 0.05 times the working pressure to ensure sufficient driving force to overcome flow resistance. The time-varying characteristics of the pressure difference are obtained through continuous monitoring, reflecting the dynamic changes in the pressure field. The spatially distributed pressure difference gradient indicates the direction of the most drastic pressure change; a larger gradient indicates stronger driving potential. The diagonal elements of the pressure difference matrix are zero, while the off-diagonal elements record the pressure difference between all possible zone pairs. Statistical analysis determines the probability distribution characteristics of the pressure differential; the mean reflects the average driving capacity, and the variance represents the stability of the pressure differential. Extreme pressure differentials identify high-energy transfer pathways that require special attention. The pressure differential analysis results not only quantify the driving potential energy but also reveal the most valuable energy transfer pathways.

[0063] Based on the calculated pressure differential distribution, the natural flow direction under pressure-driven conditions is determined using fluid mechanics principles. The basic principle for flow direction is to flow from the high-pressure area to the low-pressure area, forming a natural release path of pressure potential energy. The direction vector is calculated as F_ij = -∇P×A_eff, where ∇P is the pressure gradient, A_eff is the effective flow area, and the negative sign indicates flow along the direction of decreasing pressure. Streamline tracing starts from the center of the high-pressure area and proceeds along the direction of maximum decrease in local pressure gradient. The length of each step is determined by the local flow velocity, and tracing terminates at the boundary of the low-pressure area. Path curvature constraints ensure smooth changes in flow direction, with a curvature radius not less than 5 times the pipe diameter, avoiding local losses and flow separation caused by sharp turns. Flow distribution in multi-path cases follows the principle of minimum resistance, with the flow rate of each path proportional to its conductance. Flow around obstacles is handled using potential flow theory, with the velocity potential function satisfying the Laplace equation and the boundary condition of no penetration at the obstacle surface. Boundary layer effects lead to a decrease in near-wall velocity, and a logarithmic velocity distribution is used to correct the velocity in the mainstream region. Three-dimensional flow is simplified into two main components, horizontal and vertical, through orthogonal decomposition. Determining the flow direction field provides theoretical guidance for the design of auxiliary channels.

[0064] Based on the determined natural direction of pressure flow, a flow regulation strategy is designed between pipes. The regulation strategy does not involve physically adding new channels, but rather establishing a "virtual connection" effect between pipes through intelligent valve opening coordination. Specifically, this is achieved by simultaneously adjusting the openings of multiple pipes: when a pipe is detected to have excessively high pressure, its opening is slightly reduced while the openings of other pipes are increased accordingly, maintaining a constant total flow rate while redistributing pressure. Auxiliary channels provide dedicated paths for natural pressure flow, and their design directly impacts the synergistic effect. Channel path planning employs an improved optimal path algorithm to search for the optimal path in three-dimensional space, with the heuristic function considering both distance and pressure drop. Path constraints include avoiding major equipment, not crossing load-bearing structures, and maintaining a safe distance from existing pipelines. Channel dimensions are determined based on flow continuity; the design flow rate is based on pressure differential and time requirements, allowing flow velocity to be limited within the economic flow rate range, thereby calculating the required cross-sectional area. The hydraulic diameter is optimized by selecting from circular, square, and elliptical cross-sections, comprehensively considering flow resistance, manufacturing costs, and space constraints. The branching and merging design follows the principle of momentum conservation, with branching angles not exceeding 45° and merging using Y-shaped or arc-shaped transitions. Channel materials are selected based on fluid characteristics, pressure rating, and service life requirements, commonly using stainless steel, carbon steel, or composite materials. Support and fixation consider thermal expansion and contraction and vibration effects, employing a combination of elastic supports and guide brackets. The optimized layout of the auxiliary channel system achieves efficient connectivity between high and low pressure areas.

[0065] Through optimized adjustment strategies, a complete complementary and coordinated flow field scheme is formed by coordinating the opening control of each pipe. The core of the scheme is to establish a "virtual complementary network" between pipes: sharing the pressure status of each pipe in real time to form a pressure sensing matrix; calculating the optimal opening allocation of each pipe based on the pressure sensing matrix; and synchronously adjusting the opening of multiple pipes through a coordinated control algorithm to achieve dynamic pressure balance. The scheme integration is not a simple channel superposition, but a systematic flow field reconstruction. The flow distribution optimization establishes the objective function min Σ(f_i×L_i×ΔP_i), where f_i is the friction coefficient of the i-th segment, L_i is the pipe length, and ΔP_i is the pressure drop. The optimal flow distribution of each channel is solved through linear programming. Constraints include node flow balance, upper and lower pressure limits, and flow velocity not exceeding the maximum allowable value. The selection and characteristic matching of control valves consider the flow adjustment range, pressure drop characteristics, and response speed. Valve placement follows the proximity principle, prioritizing branch points and confluence points. The directionality of check valves ensures that the flow follows the designed path and prevents pressure backflow. Monitoring instruments, including pressure transmitters, flow meters, and thermometers, are deployed at key nodes to form a monitoring network. The startup procedure is designed with three phases: preheating, low-flow trial operation, and normal operation. Emergency response plans cover abnormal conditions such as overpressure, leakage, and blockage. The implementation of the collaborative solution transforms previously independent flow paths into a mutually supportive integrated system.

[0066] A constructive flow field complementary and synergistic scheme is adopted to generate an implementable pressure balance control strategy. The control objective function is set as J = w_1×(P_max-P_min) + w_2×E_total + w_3×Σ|dP / dt|, where J is the total objective function value, w_1, w_2, and w_3 are weighting coefficients, P_max and P_min are the maximum and minimum pressures in the system, E_total is the total energy consumption of the system, dP / dt is the pressure change rate at each monitoring point, and Σ represents the summation over all monitoring points. The first term (P_max-P_min) minimizes the pressure range, achieving uniform pressure distribution; the second term E_total reduces total energy consumption, improving operational economy; and the third term Σ|dP / dt| suppresses pressure fluctuations, enhancing system stability. The weighting coefficients are adjusted according to the importance of the operating conditions, with typical values ​​of w_1=0.5, w_2=0.3, and w_3=0.2. The control variables include the valve opening degree of each channel, pump speed, and auxiliary channel engagement status. The model predictive control algorithm solves the optimization problem at each sampling time. The prediction time domain is set to 10 times the control cycle to cover the main dynamic processes, while the control time domain is set to 3 times the control cycle to balance computational load and control performance. The state-space model is obtained through system identification and includes pressure dynamics, flow coupling, and time delay characteristics. Soft constraint techniques are used for constraint handling, allowing short-term violations of non-critical constraints. Feedback correction compensates for model errors and unmodeled dynamics. The hierarchical control architecture decomposes the complex problem, with the upper layer optimizing the steady-state operating point and the lower layer tracking the dynamic trajectory. Anomaly detection is based on residual analysis and statistical process control. Fault tolerance maintains basic functionality through control reconfiguration. A pressure balance control strategy based on a complementary flow field cooperative scheme enables intelligent pressure management of multi-bore valve systems.

[0067] Step S150: Construct a multi-valve collaborative control timing sequence based on water hammer cancellation control command and pressure balance control strategy. Analyze the pressure wave propagation path through the multi-valve collaborative control timing sequence to determine the pressure wave superposition elimination point. Extract pressure pulse resonance features from the pressure wave superposition elimination point and generate timing control parameters based on the pressure pulse resonance features.

[0068] Specifically, a multi-valve collaborative control sequence is constructed by integrating water hammer cancellation control commands and pressure balance control strategies. The water hammer cancellation control commands include the action time, pressure value, and duration of each control point, while the pressure balance control strategy provides optimized trajectories for control variables such as valve opening and pump speed. The core of the sequence construction is to coordinate the two types of control actions, avoiding mutual interference and maximizing synergistic effects. The time axis uses millisecond-level precision, with the control system's sampling period of 100ms as the basic time unit. Priority allocation assigns high priority to the water hammer cancellation command, executing it at critical moments, while pressure balance adjustment continues as background control. A conflict detection algorithm identifies duplicate commands from the same actuator and resolves them through time window merging or staggered execution. Synergistic enhancement design incorporates pressure balance pre-adjustment and post-adjustment before and after the water hammer cancellation action to enhance control effectiveness. The spatial distribution of multiple valves considers upstream and downstream relationships, with upstream valves acting earlier than downstream valves, and propagation delay calculated based on pipe length and wave velocity. A hybrid parallel and serial orchestration allows for parallel control of independent pipes and serial coordination of coupled pipes. The timing-tolerance design inserts buffer time between critical actions to prevent timing chaos caused by execution delays. The completed multi-valve collaborative control timing includes the precise timing arrangement and execution order of all control actions.

[0069] By constructing a multi-valve collaborative control timing sequence, the propagation path and interaction of pressure waves in the pipeline network are analyzed. Each control action generates pressure disturbances, which propagate in the pipeline as pressure waves. Propagation path analysis starts from the location of each valve and traces the trajectory of the pressure waves along the pipeline. Wave velocity calculation considers fluid characteristics and pipeline elasticity, with typical values ​​in the range of 1000-1400 m / s. Identification of multi-source pressure waves is achieved by determining the action time of each valve through a timing table and calculating the pressure waves generated by each source. Branching and merging during propagation occur at nodes such as tees and crosses, following the principles of acoustic propagation. Reflection phenomena occur at the pipeline end, valve closed positions, and abrupt changes in pipe diameter; the reflection coefficient is determined by impedance matching. Superposition effect analysis adopts the principle of linear superposition; when multiple pressure waves meet at the same spatiotemporal point, the pressure amplitudes are algebraically added. The cancellation condition is determined when two equal-amplitude, opposite-phase waves meet, resulting in complete cancellation with a phase difference close to 180°. Spatiotemporal scanning is performed through a gridded pipeline network, calculating the pressure superposition results at different times at each grid point. The elimination effect is evaluated by quantifying the ratio of the superimposed pressure amplitude to the original amplitude. The pressure wave superposition elimination points are concentrated in specific spatial locations and time windows, where pressure pulsations are effectively suppressed.

[0070] In some embodiments, extracting pressure pulse resonance features from the pressure wave superposition elimination point includes: monitoring the pressure pulse frequency based on the pressure wave superposition elimination point; comparing the pressure pulse frequency with the system's natural frequency to obtain a frequency matching degree; identifying the resonance enhancement interval based on the frequency matching degree; and extracting amplitude and phase information from the resonance enhancement interval as pressure pulse resonance features.

[0071] Based on the location information of the pressure wave superposition elimination points, a pressure pulse frequency monitoring scheme was set up. Monitoring points were selected at locations where the superposition elimination effect was significant but residual pulsations remained, as these locations were most sensitive to frequency characteristics. High-speed sampling was used for data acquisition; the sampling theorem requires the sampling frequency to be at least twice the highest frequency of interest, but it was actually set to 10 times to improve accuracy. Filtering preprocessing removed DC components and low-frequency drift, retaining pulse components from 1Hz to 500Hz. The frequency extraction algorithm employed an adaptive notch filter bank to track changes in the main frequency components in real time. Instantaneous frequency calculation obtained an analytical representation of the signal through Hilbert transform, deriving the instantaneous frequency from the phase derivative. Frequency stability assessment calculated the frequency standard deviation within a certain time window; stable frequency components were more likely to cause resonance. The spatial distribution of multiple frequencies reflected the differences in pulse characteristics across different regions. Frequency evolution tracking recorded the changing trends of the main frequency components over time. Abnormal frequency detection identified suddenly appearing or disappearing frequency components. The monitored pressure pulse frequency spectrum provided fundamental data for resonance analysis.

[0072] The monitored pressure pulse frequencies are compared with predetermined system natural frequencies. The system natural frequency database contains characteristic frequencies of pipe acoustic modes, structural vibration modes, and fluid-structure interaction modes. The frequency matching degree is calculated using the formula M = 1 - |f_pulse - f_natural| / f_natural, where f_pulse is the pulse frequency, f_natural is the closest natural frequency, and M close to 1 indicates a high degree of matching. Multimodal matching considers the case where a pulse frequency may be close to multiple natural frequencies simultaneously. Weight allocation is determined based on the participation factors of each mode; modes with larger participation factors have a more significant impact. Frequency drift compensation considers the influence of operating conditions such as temperature and pressure on the natural frequencies and corrects the natural frequency values ​​in real time. The time-varying characteristics of the matching degree are obtained through sliding window analysis, with a window width of 1 second and an update period of 100ms. Statistical distribution analysis is used to analyze the probability density function of the matching degree to identify the frequency range most likely to resonate. A threshold setting marks cases with a matching degree greater than 0.9 as potential resonances.

[0073] For example, identifying the resonance enhancement interval based on the frequency matching degree includes: evaluating a frequency proximity coefficient based on the frequency matching degree; predicting an amplitude amplification factor based on the proximity coefficient; determining a resonance start point based on the moment when the amplitude amplification factor exceeds a threshold; and marking the time from the resonance start point to the moment when the amplitude amplification factor falls back as the resonance enhancement interval.

[0074] First, a more refined frequency proximity coefficient is calculated based on the frequency matching degree M. The proximity coefficient K = M² / (1-M²+ε) undergoes a nonlinear transformation, where ε=0.01 to prevent the denominator from being zero. This transformation amplifies differences in high matching degrees. For multi-frequency superposition cases, vector synthesis is used, with K_total = √(ΣK_i²) considering the combined effects of multiple frequencies. A temperature correction coefficient compensates for the K value based on real-time temperature, increasing by 2% for every 10°C increase in temperature. Pressure correction similarly considers the impact of operating pressure on resonance characteristics. Historical data fitting establishes an empirical relationship between the K value and the actual resonance intensity. Confidence interval calculation provides the uncertainty range of the K value for risk assessment. Time smoothing eliminates rapid fluctuations in the K value through moving averages. Spatial interpolation estimates the distribution of the K value among monitoring points. Next, based on the proximity coefficient K, the amplitude amplification factor under resonance conditions is predicted. The prediction model is based on the response characteristics of a single-degree-of-freedom resonant system, with an amplification factor A = 1 / √((1-K)² +(2ζK)²), where ζ is the system damping ratio. The damping ratio is estimated through free vibration decay tests or the half-power bandwidth method, with typical values ​​ranging from 0.02 to 0.05. Nonlinear correction is considered for large amplitudes, A_corrected = A×(1-β×A²), where β is the nonlinear coefficient. Multimodal coupling is handled using the modal superposition method, with the total amplification factor being the vector sum of the contributions from each mode. Confidence assessment provides the reliability of the prediction based on historical prediction error statistics. Parameter sensitivity analysis identifies key factors affecting prediction accuracy. Real-time correction adjusts model parameters based on the deviation between the measured amplitude and the predicted value. The prediction time domain is extended to the next 10 sampling periods to provide an early warning time. Then, based on the predicted amplitude amplification factor, the resonance initiation time is determined. The resonance initiation point is determined using a dual criterion: the amplification factor A exceeds a set threshold A_th=3 and the growth rate dA / dt>0. The threshold is adaptively adjusted based on the background vibration level, A_th = 3 + 2×σ_background, where σ_background is the standard deviation of the background vibration. Precise location of the starting point is achieved through a reverse search, finding the moment when A first exceeds 0.9×A_th. The resonance development process is divided into three stages: a rapid growth period, a quasi-steady-state period, and a decay period. Peak identification focuses not only on the maximum value but also records the second-highest value for mode analysis. Duration estimation is based on the system's quality factor Q and excitation characteristics. Resonance termination is determined when A falls below 1.5 for three consecutive sampling periods. Finally, the period from the resonance initiation point to the point where the amplitude amplification factor falls back is marked as the resonance enhancement interval. This is achieved by continuously monitoring the amplitude amplification factor A(t). When A(t) monotonically decreases from its peak and maintains a downward trend for more than five consecutive sampling periods, decay begins. The interval termination point is defined as the first moment when A(t) falls below the threshold A_th, and A(t) must remain below the threshold for the next three sampling periods to confirm stability.The interval duration is the time difference between the end and start times, typically ranging from half a second to three seconds. The energy integral within the interval, E_resonance = ∫[t_start to t_end] A²(t)dt, quantifies the overall level of resonance intensity. Multiple interval processing is used when a single excitation produces multiple resonance peaks; each is labeled as a sub-interval and its characteristics are calculated. Interval overlap detection prevents confusion between adjacent resonance intervals; overlapping portions are assigned to the interval with stronger energy. The interval database records all identified resonance enhancement intervals and their characteristic parameters.

[0075] From the identified resonance enhancement intervals, amplitude and phase information are extracted as complete resonance features. Amplitude feature extraction includes peak amplitude, average amplitude, amplitude envelope, and amplitude modulation depth. Peak amplitude is directly read as the maximum value within the interval, and average amplitude is obtained through integral averaging. The amplitude envelope is extracted using Hilbert transform, reflecting the slow changes in amplitude. Phase features focus on initial phase, phase drift rate, and phase jump. Initial phase is determined by the signal phase at the start of the interval. Phase drift is obtained by linearly fitting the slope of the phase-time curve. Instantaneous features include the joint distribution of instantaneous frequency and instantaneous amplitude. Statistical features calculate the higher-order moments of the amplitude distribution to assess the stability of the resonance. Spatial coherence analysis examines the phase relationship of resonances at different locations. Time-frequency features obtain the time-frequency energy distribution of the resonance through wavelet transform. Feature vector construction combines all features into a high-dimensional vector for pattern recognition.

[0076] Based on the extracted pressure pulse resonance characteristics, optimized timing control parameters are generated. The goal of parameter generation is to suppress harmful resonances and utilize beneficial resonances to achieve stable and efficient system operation. Suppression strategies target resonances with excessive amplitude by adjusting the control timing to disrupt resonance conditions. A timing offset Δt = T_resonance / 4 ensures a 90° phase difference between the excitation and response, where T_resonance is the resonance period. Frequency modulation fine-tunes the valve actuation rate to deviate the excitation frequency from the resonance frequency by more than 5%. Amplitude modulation reduces the control intensity in the resonance range, decreasing energy input. Phase control introduces a specific phase difference between multiple valves, causing the pressure waves generated by each valve to mutually suppress each other. A strategy is used to identify beneficial low-amplitude resonances, enhancing transmission efficiency through synchronous control. Parameter optimization employs a genetic algorithm, with the objective function including resonance suppression effect and control energy consumption. Constraints ensure parameters remain within the actuator's capability range. Robust design considers parameter uncertainties, maintaining performance near nominal values. The generated parameter table includes the actuation time correction, rate adjustment coefficient, and amplitude scaling factor for each valve. The final generated timing control parameters achieve proactive management of pressure pulse resonances.

[0077] Step S160: The timing control parameters are matched with the cascade action chain to generate synchronous control nodes. The action time of each flow path is determined based on the synchronous control nodes. A time-sharing control sequence is generated according to the action time to complete the intelligent coordinated switching control of the flow path of the multi-flow path valve.

[0078] In some embodiments, the step of timing-matching the timing control parameters with the cascaded action chain to generate a synchronization control node includes: performing delay characteristic analysis on the timing control parameters to identify fixed delay components and variable delay components; comparing the fixed delay components and the variable delay components with the cascaded action chain to obtain timing deviations; performing compensation calculations based on the timing deviations to determine the optimal synchronization time; and marking the control point corresponding to the optimal synchronization time as a synchronization control node.

[0079] A thorough analysis of the delay characteristics of the timing control parameters is conducted to distinguish different types of delay components. The obtained timing control parameters include a pre-compensation for system delay in the action timing correction, and the sufficiency and accuracy of this compensation need to be analyzed. Fixed delay component analysis identifies a constant compensation portion from the correction, corresponding to the physical propagation time t_prop = L / a (where t_prop is the propagation time, L is the propagation distance, and a is the speed of sound of the pressure wave), the actuator's inherent delay (typically 20-50ms), and the signal processing delay (approximately 10ms). The rationality of the compensation is evaluated by comparing the correction in the parameters with theoretically calculated values. Variable delay component analysis examines the dynamic adjustment portion of the parameters; the rate adjustment coefficient affects the action completion time, thus generating variable delay. When the rate coefficient is less than 1, the action slows down, and the equivalent delay increases by Δt_var = (1 / k_rate - 1)×t_nominal, where k_rate is the rate adjustment coefficient and t_nominal is the nominal action time. While the amplitude scaling factor primarily affects the control strength, it also indirectly affects the response time by changing the valve stroke. Parameter coupling effect analysis reveals the comprehensive delay characteristics when multiple control parameters act simultaneously, showcasing the mutual influence between parameters. Statistical feature extraction extracts the distribution patterns of parameters under different operating conditions; the fixed compensation portion remains stable, while the variable portion exhibits a normal distribution. Through delay characteristic analysis, both considered and insufficiently considered delay factors in the control parameters are identified.

[0080] The identified fixed and variable delay components are compared in detail with the timing requirements in the cascaded action chain. The cascaded action chain defines the action timing under ideal conditions, including the time intervals between nodes and causal dependencies. The timing construction process transforms abstract action nodes into specific time arrangements, with the time intervals between nodes determined by the difference in trigger times, and causal dependencies reflected by the sequential order of nodes. The comparison method uses timing diagram alignment, superimposing the actual delays onto the ideal timing. For fixed delay comparison, the offset Δt_fixed = t_actual_fixed - t_ideal is directly calculated, where t_actual_fixed is the actual time including the fixed delay, and t_ideal is the ideal design time. For variable delay comparison, its randomness is considered, and the expected deviation and variance are calculated. Cumulative effect analysis is used to analyze the propagation and amplification of delays in multi-level cascading. The critical path is identified to determine the control path with the greatest delay impact. Timing margin assessment evaluates whether the time margin reserved in the original design is sufficient to cover the actual delay. Deviation types are classified, including constant offset, proportional deviation, and random jitter. Spatial distribution analysis is used to analyze delay differences at different locations, with the delay at the far end of the path typically being larger. The time-varying characteristics take into account the drift trend of delay over time. The comparison results form a deviation matrix, recording the timing deviation of each node under different operating conditions.

[0081] Based on the acquired timing deviation, system compensation calculations are performed to determine the optimal synchronization time to eliminate or minimize the deviation. The compensation strategy employs a combination of feedforward and feedback. Feedforward compensation is based on a known fixed delay, while feedback compensation addresses variable delays. Fixed delay compensation directly advances the corresponding time, with the compensated time t_comp = t_orig - Δt_fixed, where t_orig is the original time. Variable delay compensation uses statistical prediction: t_comp = t_orig - (μ_var + k × σ_var), where μ_var is the mean of the variable delay, σ_var is the standard deviation, and k is the confidence coefficient, with a value of 1.96 corresponding to a 95% confidence level. Multi-node synchronization uses the least squares method, with the objective function being min Σ(t_i_actual - t_i_target)², minimizing the total deviation by adjusting the times of each node. Constraints include maintaining causality and execution capability limitations. Robust design ensures that synchronization effectiveness is maintained within a delay variation range of ±20%. Sensitivity analysis identifies the parameters with the greatest impact on synchronization effectiveness. The result of the compensation calculation is a set of optimized control times that fully take into account the effects of various delay factors.

[0082] The calculated optimal synchronization moments are mapped to the control system and marked as synchronization control nodes. The marking process first locates each optimal synchronization moment on the time axis with an accuracy of 1ms. Node attribute definitions include key information such as node ID, timestamp, participating path list, and synchronization type. Synchronization types are divided into strong synchronization and weak synchronization. Strong synchronization requires that the time deviation of each path's action does not exceed ±5ms, while weak synchronization allows a deviation of ±20ms. Spatial mapping associates time nodes with specific control equipment and measurement point locations. Priority assignment is based on the node's impact on system performance; pressure balance-related nodes receive high priority, while auxiliary function nodes receive low priority. Trigger conditions include two modes: time-triggered and event-triggered. Normally, time-triggered events are used, while special operating conditions can be triggered by events. Node association establishes logical relationships between synchronization nodes, including predecessor, successor, and mutual exclusion relationships. Visualization uses a Gantt chart to intuitively display the distribution of synchronization nodes on the time axis. Through a systematic marking process, a complete set of synchronization control nodes is formed, with each node containing precise time and spatial information.

[0083] Based on the generated synchronous control nodes, the action timing of each path in coordinated control is accurately determined. The synchronous control nodes define time anchor points for the coordinated actions of multiple paths, and the specific action timing of each path is arranged around these anchor points. Timing calculation first determines the absolute time of each synchronous node, establishing a unified time coordinate with the system startup time as the zero point. Path priorities are determined based on fluid dynamics characteristics and control objectives, with the main path undertaking the primary flow regulation task given priority, and auxiliary paths coordinating with the main path's actions. Propagation delay compensation is calculated based on the physical length and pressure wave velocity of each path, ensuring that downstream actions are executed after the upstream influence arrives. Parallel actions identify actions that can be executed simultaneously under the same synchronous node, improving control efficiency. Serial constraints ensure that actions with causal dependencies are executed in the correct order, with subsequent actions only starting after the preceding action is completed. The action window sets an allowed execution time range of [t_min, t_max] for each action, within which it can be fine-tuned according to real-time status. A safety interval maintains a time interval of at least 50ms between adjacent actions to prevent action overlap and interference. Special timing arrangements are designed for special operating conditions such as system startup, normal operation, and operating condition switching. Determining the timing of actions for each path transforms the abstract synchronization requirement into a concrete execution schedule.

[0084] A time-sharing control sequence is generated based on the determined action times to achieve intelligent coordinated switching control of multi-port valve flow channels. Time-sharing control decomposes the continuous control process into discrete time slices, each with a length of 100ms, consistent with the sampling period of the control system. The sequence encoding uses a structured format, with each instruction containing five elements: timestamp, port identifier, action type, target value, and duration. Action types cover basic control functions such as valve opening adjustment, flow setting, and pressure regulation. The target value is calculated based on the amplitude scaling factor of S150 and the current operating conditions to ensure moderate control action intensity. A gradual control strategy decomposes large-amplitude adjustments into multiple small steps, with each step changing no more than 10%, avoiding system shock. Parallel instructions allow control commands to be issued simultaneously to multiple ports within the same time slice, improving response speed. Conditional judgment logic determines subsequent actions based on the execution result of the previous time slice, achieving closed-loop control. Sequence integrity is ensured through beginning and end markings and checksums to guarantee error-free transmission. Control commands are sent to each actuator via the fieldbus, and the actuators execute them precisely according to the timestamp, ultimately completing the intelligent coordinated switching control of the flow channels of multi-port valves.

[0085] To implement the intelligent coordinated switching control method for multi-bore valve flow channels corresponding to the above method embodiments, and to achieve the corresponding functions and technical effects. See also Figure 2 , Figure 2 This diagram illustrates a structural block diagram of an intelligent coordinated switching control system 200 for multi-bore valve flow channels according to an embodiment of this application. For ease of explanation, only the parts relevant to this embodiment are shown. The intelligent coordinated switching control system 200 for multi-bore valve flow channels provided in this embodiment includes:

[0086] The flow state identification module 201 is used to collect real-time fluid parameters of multi-bore valves, perform feature analysis on the real-time fluid parameters to identify the current flow channel working mode, and construct a flow channel state transition matrix using the flow channel working mode;

[0087] Path planning module 202 is used to generate a candidate path set by performing reverse path deduction through the flow channel state transition matrix, perform pressure wave time series analysis on the candidate path set to identify leverage switching points, generate a time series leverage switching sequence based on the leverage switching points, and construct a cascaded action chain based on the time series leverage switching sequence.

[0088] Water hammer control module 203 is used to extract the pressure gradient distribution and flow velocity change curve of each diameter from the cascade action chain, predict the water hammer intensity distribution based on the pressure gradient distribution, determine the water hammer occurrence sequence through the flow velocity change curve, design reverse pressure wave parameters by combining the water hammer intensity distribution and the water hammer occurrence sequence, and generate water hammer cancellation control commands using the reverse pressure wave parameters.

[0089] The collaborative control module 204 is used to acquire the pressure distribution characteristics of each path in the cascade action chain, identify high-pressure and low-pressure areas based on the pressure distribution characteristics, construct a flow field complementary collaborative scheme based on the distribution of the high-pressure and low-pressure areas, and generate a pressure balance control strategy using the flow field complementary collaborative scheme.

[0090] The timing optimization module 205 is used to construct a multi-valve collaborative control timing based on the water hammer cancellation control command and the pressure balance control strategy, analyze the pressure wave propagation path through the multi-valve collaborative control timing to determine the pressure wave superposition elimination point, extract pressure pulse resonance features from the pressure wave superposition elimination point, and generate timing control parameters based on the pressure pulse resonance features.

[0091] The execution control module 206 is used to perform timing matching between the timing control parameters and the cascade action chain to generate synchronous control nodes, determine the action time of each flow path based on the synchronous control nodes, generate a time-sharing control sequence according to the action time, and complete the intelligent coordinated switching control of the flow path of the multi-flow-path valve.

[0092] The purpose of the above embodiments is to reproduce and derive the technical solution of the present invention by way of example, and to fully describe the technical solution, purpose and effect of the present invention. The purpose is to enable the public to have a more thorough and comprehensive understanding of the disclosure of the present invention, and not to limit the scope of protection of the present invention.

[0093] The above embodiments are not an exhaustive list based on the present invention, and there may be many other embodiments not listed. Any substitutions and improvements made without departing from the concept of the present invention are within the protection scope of the present invention.

Claims

1. A method for intelligent coordinated switching control of flow channels in a multi-bore valve, characterized in that, include: Real-time fluid parameters of multi-bore valves are collected, and feature analysis is performed on the real-time fluid parameters to identify the current flow channel working mode. The flow channel working mode is then used to construct a flow channel state transition matrix. The candidate path set is generated by reverse path deduction through the flow channel state transition matrix. Pressure wave time series analysis is performed on the candidate path set to identify leverage switching points. A time series leverage switching sequence is generated based on the leverage switching points. A cascaded action chain is constructed based on the time series leverage switching sequence. The pressure gradient distribution and velocity change curve of each path are extracted from the cascade action chain. The water hammer intensity distribution is predicted based on the pressure gradient distribution. The water hammer occurrence sequence is determined through the velocity change curve. The reverse pressure wave parameters are designed by combining the water hammer intensity distribution and the water hammer occurrence sequence. The water hammer cancellation control command is generated using the reverse pressure wave parameters. Obtain the pressure distribution characteristics of each path in the cascade action chain, identify high-pressure and low-pressure areas based on the pressure distribution characteristics, construct a flow field complementary and cooperative scheme based on the distribution of the high-pressure and low-pressure areas, and generate a pressure balance control strategy using the flow field complementary and cooperative scheme. Based on the water hammer cancellation control command and the pressure balance control strategy, a multi-valve collaborative control timing sequence is constructed. Pressure wave propagation path analysis is performed through the multi-valve collaborative control timing sequence to determine the pressure wave superposition elimination point. Pressure pulse resonance features are extracted from the pressure wave superposition elimination point. Timing control parameters are generated based on the pressure pulse resonance features. The timing control parameters are matched with the cascaded action chain to generate synchronous control nodes. The action time of each flow path is determined based on the synchronous control nodes. A time-sharing control sequence is generated according to the action time to complete the intelligent coordinated switching control of the flow path of the multi-flow path valve.

2. The method according to claim 1, characterized in that, The process of generating a candidate path set through reverse path deduction using the flow channel state transition matrix includes: Extract the switching cost between states based on the aforementioned flow channel state transition matrix; The total path cost is obtained by summing the switching cost from the target state in reverse order. Based on the total cost of the path, multiple paths with the lowest cost are selected to form a candidate path set.

3. The method according to claim 1, characterized in that, The step of performing pressure wave time series analysis on the candidate path set to identify leveraged switching points includes: The pressure wave propagation characteristics of each path are extracted based on the candidate path set. Time-series energy analysis was performed on the propagation characteristics of the pressure wave to obtain the energy overlap region; The energy overlap region is used to determine the point where power can be switched.

4. The method according to claim 1, characterized in that, The construction of the cascaded action chain based on the time-series leveraging switching sequence includes: The remaining energy value of each switching point is extracted based on the time-series leveraging switching sequence; The available energy window is identified based on the remaining energy value; Determine the optimal triggering time based on the energy window; The timing-based switching sequences are linked together to form a cascaded action chain at the optimal triggering time.

5. The method according to claim 1, characterized in that, The design of reverse pressure wave parameters, combining the water hammer intensity distribution and the water hammer occurrence timing, includes: Peak pressure amplitude is extracted based on the water hammer intensity distribution; The arrival time of the pressure wave is determined based on the water hammer occurrence sequence. The required reverse pressure value is calculated based on the peak pressure amplitude. The required reverse pressure value is matched with the arrival time of the pressure wave to generate reverse pressure wave parameters.

6. The method according to claim 1, characterized in that, The step of constructing a complementary and coordinated flow field scheme based on the distribution of the high-pressure region and the low-pressure region includes: Based on the distribution of the high-pressure zone and the low-pressure zone, the pressure difference value between each region is extracted; The natural direction of pressure flow is determined based on the pressure difference value; The auxiliary channel is designed based on the natural direction of the pressure flow to obtain an optimized flow channel layout; The optimized flow channel layout forms a complementary and synergistic flow field scheme.

7. The method according to claim 1, characterized in that, The extraction of pressure pulse resonance features from the pressure wave superposition elimination point includes: Based on the pressure wave superposition elimination point, monitor the pressure pulse frequency; The frequency matching degree is obtained by comparing the frequency of the pressure pulse with the system's natural frequency. The resonance enhancement region is identified based on the frequency matching degree; Amplitude and phase information are extracted from the resonance enhancement region as pressure pulse resonance features.

8. The method according to claim 1, characterized in that, The step of performing timing matching between the timing control parameters and the cascaded action chain to generate a synchronization control node includes: Delay characteristic analysis is performed on the timing control parameters to identify fixed delay components and variable delay components; The fixed delay component and the variable delay component are compared with the cascaded action chain to obtain the timing deviation; Based on the timing deviation, compensation calculations are performed to determine the optimal synchronization time; The control point corresponding to the optimal synchronization time is marked as a synchronization control node.

9. The method according to claim 7, characterized in that, The step of identifying the resonance enhancement region based on the frequency matching degree includes: The frequency similarity coefficient is evaluated based on the frequency matching degree. The amplitude amplification factor is predicted based on the proximity coefficient. The resonance initiation point is determined based on the moment when the amplitude amplification factor exceeds the threshold. The period from the resonance initiation point to the point where the amplitude amplification factor drops back is marked as the resonance enhancement interval.

10. An intelligent collaborative switching control system for multi-bore valve flow channels, characterized in that, include: The flow state identification module is used to collect real-time fluid parameters of multi-bore valves, perform feature analysis on the real-time fluid parameters to identify the current flow channel working mode, and construct a flow channel state transition matrix using the flow channel working mode; The path planning module is used to generate a candidate path set by performing reverse path deduction through the flow channel state transition matrix, perform pressure wave time series analysis on the candidate path set to identify leverage switching points, generate a time series leverage switching sequence based on the leverage switching points, and construct a cascaded action chain based on the time series leverage switching sequence. The water hammer control module is used to extract the pressure gradient distribution and flow velocity change curve of each diameter from the cascade action chain, predict the water hammer intensity distribution based on the pressure gradient distribution, determine the water hammer occurrence sequence through the flow velocity change curve, design reverse pressure wave parameters by combining the water hammer intensity distribution and the water hammer occurrence sequence, and generate water hammer cancellation control commands using the reverse pressure wave parameters. The collaborative control module is used to acquire the pressure distribution characteristics of each path in the cascade action chain, identify high-pressure and low-pressure areas based on the pressure distribution characteristics, construct a flow field complementary collaborative scheme based on the distribution of the high-pressure and low-pressure areas, and generate a pressure balance control strategy using the flow field complementary collaborative scheme. The timing optimization module is used to construct a multi-valve collaborative control timing based on the water hammer cancellation control command and the pressure balance control strategy, analyze the pressure wave propagation path through the multi-valve collaborative control timing to determine the pressure wave superposition elimination point, extract pressure pulse resonance features from the pressure wave superposition elimination point, and generate timing control parameters based on the pressure pulse resonance features. The execution control module is used to perform timing matching between the timing control parameters and the cascaded action chain to generate synchronous control nodes, determine the action time of each flow path based on the synchronous control nodes, generate a time-sharing control sequence according to the action time, and complete the intelligent coordinated switching control of the flow path of the multi-flow-path valve.

Images