Grid
Load and generation forecasting
What a forecast leaned on, hour by hour, in megawatts: yesterday's load, the hour, the weather.
Solutions · Energy and industry
Load, output and failure models work until the conditions change. Kaimei shows what each forecast leaned on, in its own units, and flags the hour the inputs leave the range the model was trained in.
Where it fits
Grid
What a forecast leaned on, hour by hour, in megawatts: yesterday's load, the hour, the weather.
Assets
Why this unit was flagged, from its sensor features, and which sensor has drifted since the model was fitted.
Quality
Grad-CAM and occlusion on a defect classifier, with a check that the model looked at the part and not the background.
Process
Behaviour curves show how predicted output responds to each setting, and whether that response is monotonic.
Field
A sensor that starts reporting in another unit shows up as input drift within one batch.
Sites
Run Kaimei on your own infrastructure, or upload shapes only and keep the weights on site.
A worked example
Public dataPower Consumption of Tetouan City, UCI Machine Learning Repository (Salam & El Hibaoui, 2018).load-forecastscikit-learn · neural network
A small neural network, two hidden layers behind a scaler, forecasts hourly city load in megawatts from eight inputs: hour, weekend, four weather readings and the load a day and a week earlier. It is trained on January to June 2017, 4,176 hours, and then meets July.
POST https://api.kaimei.ai/v1/explanations
One forecast
At 21:00 on 30 June the model forecasts 105.5 MW against an actual 104.4. The explanation starts from the 70.6 MW average and adds +20.3 for yesterday's load and +12.9 for last week's.
At 11:00 on 24 July it forecasts 87.2 MW and the city draws 104.0. The explanation shows why: it is 29.4 °C, and temperature adds only +0.7 MW. The model leans on a lag of 88.6 MW (+9.5) because nothing in its training months taught it what that heat does to demand.
Out of range
Kaimei compares the latest 500 logged hours with the reference hours, one input at a time. By late June temperature has already drifted from a reference that starts in January. Through July, wind speed and both lagged loads follow. Nobody needs to wait for the actuals to learn that the forecast is extrapolating: the alert arrives with the inputs.
temperature, wind_speed, load_168h_ago, solar_irradiance. 500 logged predictions against 4,000 reference rows.temperature, wind_speed, load_24h_ago, load_168h_ago. 500 logged predictions against 4,000 reference rows.Population stability index per input. Bars are cut off at 0.5; the line marks 0.25, where Kaimei calls drift significant.
What it learned
load_24h_ago is the most influential input, followed by load_168h_ago and hour. Together they carry 84% of the measured importance (permutation importance over 2000 rows).
Temperature holds 3.3% of measured importance. For a model that will face a summer, that is the finding: retrain with a full year, or give the forecast a temperature-driven term.
load_24h_ago increases load_mw, most sharply between 80.19 MW and 91.88 MW (range 58.4 → 86.5).temperature has a non-monotonic effect on load_mw: it bottoms out around 14.39 °C (range 68.3 → 70.1).hour has a non-monotonic effect on load_mw: it bottoms out around 4 (range 67.2 → 72.1).load_24h_ago is the most influential input, followed by load_168h_ago and hour. Together they carry 84% of the measured importance (permutation importance over 2000 rows).Inside the network
An uploaded network is drawn layer by layer, and a trace walks one prediction through it: the inputs, each weight and bias, each activation, the output. Repeated blocks fold, so a large model stays readable.
client.models.activations returns what each layer sawOutputs move by 3.8% of their spread under 5%-σ input noise. Most sensitive inputs: wind_speed, load_24h_ago.
Predictions range from 38.9 to 106 (median 70) across 2,000 reference rows.
hour has a non-monotonic effect.Who it is for
See what each forecast leaned on and how the model responds to every input, and compare next season's retrain with this one.
An alert when the inputs leave the range a model was trained in, with the input named, before the error shows up in the actuals.
Why a unit or a part was flagged, in sensor terms, and whether an inspection model looked where it should.
Coming soon
What we are building now, and what to use until each one ships.
Forecasters that read a history step by step, with each forecast attributed to time steps as well as inputs. Until then, a model on lags, rolling averages and calendar features works.
Several outputs per model, so every horizon of one forecaster is explained in a single call. Until then, register one model per horizon.
Explanations for the boxes and masks a model draws on inspection imagery, not only for its label. Image classification works today.
Drift and performance checks run by Kaimei on a schedule, with alerts sent to email, Slack, PagerDuty or a webhook. Until then, run them from your own scheduler.
Said plainly. Kaimei explains and monitors the models you already run. It does not control equipment, and it does not establish cause and effect.
What works today, and everything on the way, is in Supported models and what’s coming.
Bring a forecast or a sensor model. See what it leans on before the season turns.