The invention discloses an
energy consumption risk closed-loop monitoring method,
system, equipment and medium based on a brain-like
inference model, and belongs to the technical field of energy
digitization and intellectualization, and the method comprises the steps: collecting multi-source
energy consumption data in real time, converting the multi-source
energy consumption data into a
pulse sequence through a pulse
encoder, constructing a layered space-time
memory pool to process the
pulse sequence, and carrying out the real-time monitoring of the multi-source energy consumption data. And based on a neural
plasticity rule, carrying out dynamic weight migration on a current risk mode and a historical risk disposal strategy, analyzing a risk conduction path, outputting risk inducements, calculating risk
hormone factors, activating a dual-channel response mechanism and generating a disposal scheme by using a risk
traceability tree diagram, and optimizing scheme weights through Q-learning to generate a virtual risk scene. And rehearsing the treatment scheme effect, and dynamically updating the
model parameters according to the user response data and the
treatment effect. According to the invention, through a multi-level cooperation and closed-loop feedback mechanism, full-process intelligentization of energy consumption risk monitoring is realized, the
electric charge recovery rate is improved, and the
system robustness is enhanced.